OPINION AND ORDER

BECHTLE, District Judge.

INTRODUCTION

Plaintiff is a former Assistant Professor in the Education Program at Swarthmore College ("Swarthmore"). On February 29, 1972, plaintiff received notification from Swarthmore that she would not be reappointed as an Assistant Professor for the 1972-73 academic year due to "logistical" considerations. After exhausting the appropriate administrative remedies, plaintiff commenced this class action alleging that this nonrenewal was based solely on account of her sex, in violation of, inter alia, the Equal Employment Opportunity Act of 1972, Pub.L. No. 92-261, 86 Stat. 103 (1972), 42 U.S.C. § 2000e et seq. (Supp. V), amending, Civil Rights Act of 1964, 78 Stat. 253. (All subsequent citations to Title VII in this Opinion are to the 1964 Act as amended.) The action was bifurcated, and the liability aspect of the case was tried, non-jury, before this Court. After careful review and consideration of the testimony and exhibits presented at trial, the pre- and post-trial briefs and arguments of counsel, and the proposed findings of fact and conclusions of law submitted by the parties, the Court makes the following narrative findings of fact and conclusions of law. The bracketed references to the record set forth the primary sources from which the statements contained herein are drawn.

FINDINGS OF FACT

I. PARTIES

The only individually named plaintiff, Barbara Z. Presseisen, was employed by defendant Swarthmore as a part-time Lecturer in the academic year 1970-71, and as a full-time Assistant Professor in the academic year 1971-72.

The intervening plaintiff, the U.S. Equal Employment Opportunity Commission ("EEOC"), was granted leave to intervene as a plaintiff by Order dated June 4, 1976.

Defendant Swarthmore is an "employer" within the definition of 42 U.S.C. § 2000e(b).

Defendant Charles E. Gilbert served as the Provost of Swarthmore from the period of 1969 through 1974. The Provost is the chief academic officer under the President and has the primary responsibility for faculty personnel matters. [N.T. 23-184.]

II. BACKGROUND

On June 2, 1972, plaintiff filed a charge against defendants with the Pennsylvania Human Relations Commission ("PHRC") alleging that defendants had discriminated against her on account of sex.

On November 27, 1972, plaintiff filed a charge with the EEOC, alleging that defendants had discriminated against her on account of sex.

After appropriate deferral to the PHRC, and after investigating plaintiff's charge, the EEOC issued a right-to-sue letter to plaintiff on May 6, 1974.

Plaintiff then filed her judicial complaint on May 28, 1974.

Pursuant to plaintiff's motion to amend, this Court granted leave to plaintiff to file an amended complaint, which included, inter alia, class action allegations pursuant to Rule 23, Fed.R.Civ.P. Plaintiff's amended complaint was filed on December 11, 1974.

Pursuant to Rule 23(b)(2) of the Federal Rules of Civil Procedure, by Opinion and Order dated March 25, 1976, this Court certified this action as a class action. We defined the class as all present female faculty employees at Swarthmore College, all future female faculty employees at Swarthmore College, all former female faculty employees who left the employ of Swarthmore College on or subsequent to March 24, 1972, and those applicants for faculty positions who applied to Swarthmore College and were rejected on or subsequent to March 24, 1972, who have been, are at present being or in the future may be:

(1) denied hire, promotion and/or tenure at Swarthmore College or were not recruited by Swarthmore on account of their sex; and/or

(2) denied equal compensation for performance of substantially similar work as is performed by male faculty members at Swarthmore College, on account of their sex.

However, we certified the class action as to declaratory and injunctive relief only. A certification ruling on damages in the form of back pay was deferred until after a determination of liability.

In addition, the March 25 Opinion and Order dismissed plaintiff's claims brought under 42 U.S.C. § 1981 and 1985(3), and dismissed plaintiff's Title VII claim against those individual defendants (Theodore Friend, Stephen G. Lax and Alice K. Brodhead) not named in plaintiff's charge before the EEOC. By Order dated January 12, 1977, we also dismissed plaintiff's claim brought under 42 U.S.C. § 1983.

At the time trial commenced, the only defendants before the Court were Swarthmore and Mr. Gilbert, and plaintiffs' claims were based solely on Title VII.

III. THE TRIAL

Plaintiffs' evidence at trial consisted mainly of the testimony of 16 class members, Barbara Presseisen's own testimony as individual plaintiff, the testimony of a statistical expert witness, Dr. John deCani, Professor of Statistics at the University of Pennsylvania, and numerous exhibits. Plaintiffs included in their presentation of evidence in the class case hundreds of documents from faculty personnel files, which they contend, when considered in conjunction with the testimony and statistical evidence, establish a prima facie case of a pattern and practice of sex discrimination by defendants. The Court directed plaintiffs' counsel to file a written statement to identify specifically what women were unfairly treated due to their sex and in what way. Plaintiffs, in response, submitted "Plaintiffs' Memorandum of Comparisons."

At the close of plaintiffs' evidence, defendants moved for dismissal pursuant to Fed.R.Civ.P. 50(a), and later amended the motion to dismiss pursuant to Fed.R.Civ.P. 41(b). The Court deferred ruling on defendants' motion and requested defendants to present their evidence in defense.

Defendants' primary witness was Harold E. Pagliaro, present Provost of Swarthmore, who described all aspects of Swarthmore's employment practices and procedures regarding the faculty. He further testified with respect to defendants' response to plaintiffs' Memorandum of Comparisons. [Ex. D-147.] Defendants also presented Dr. Robert Cross, Alice Brodhead and Mr. Gilbert, all of whom testified with respect to Barbara Presseisen's individual case. Finally, defendants offered the testimony of two statistical expert witnesses, Dr. Paul Meier, Professor of Statistics at the University of Chicago, and Dr. Gudmund Iversen, Professor of Statistics at Swarthmore College.

Plaintiffs offered rebuttal testimony, and defendants surrebuttal.

DISCUSSION OF APPLICABLE LEGAL PRINCIPLES

On May 31, 1977, in International Brotherhood of Teamsters v. United States, 431 U.S. 324, 97 S.Ct. 1843, 52 L.Ed.2d 396, the Supreme Court set forth specific guidelines with respect to the burden of proof of the respective parties in a Title VII class action case. Although Teamsters was a Government-initiated pattern or practice suit, the Court finds, and all parties are in agreement, that the evidentiary bases for a private across-the-board class action suit are almost identical to a Government pattern or practice suit. Accordingly, we will look to Teamsters as our primary source of guidance in this case.

Teamsters makes clear that the plaintiff must prove illegal discrimination in his or her individual case and a pattern or practice of discrimination against the class he or she represents. Quite simply, plaintiffs in this case contend that Swarthmore regularly and purposefully treated women faculty members less favorably than male faculty members. They allege disparate treatment with respect to, inter alia, recruitment, hiring, salary, promotion, reappointments and tenure. Accordingly, as stated in Teamsters, at 335, 97 S.Ct. at 1854, "[t]he ultimate factual issues are thus simply whether there was a pattern or practice of such disparate treatment and, if so, whether the differences were `racially [or in this case sexually] premised.'" (Citation omitted.) (Footnote omitted.)

Before we address the specific burden of proof with respect to the individual and the class case, we note that the Court in Teamsters recognized that there are two types of discrimination in Title VII cases. The first are claims of disparate treatment wherein the "employer simply treats some people less favorably than others because of their race, color, religion, sex, or national origin." Teamsters, supra, at 335 n. 15, 97 S.Ct. at 1854 n. 15. The other type of claims stress disparate impact which "involve employment practices that are facially neutral in their treatment of different groups but that in fact fall more harshly on one group than another and cannot be justified by business necessity." Id. Since plaintiffs concede in their "Post Trial Brief in Support of Findings of Fact and Conclusions of Law," at pp. 2-4, that the entire thrust of their case was to establish that disparate treatment was accorded to women faculty members by Swarthmore in all terms and conditions of employment, we need only concern ourselves with the "disparate treatment" type of discrimination.

The distinction is an important one since, as was stated in Teamsters, supra, at 335 n. 15, 97 S.Ct. at 1854 n. 15, under a disparate treatment theory, "[p]roof of discriminatory motive is critical," whereas under a disparate impact theory proof of discriminatory motive is not required. Id. See, e. g., Griggs v. Duke Power Co., 401 U.S. 424, 430-432, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971); Dothard v. Rawlinson, 433 U.S. 321, 328, 97 S.Ct. 2720, 2726, 53 L.Ed.2d 786 (1977). As defendants point out in their Supplemental Memorandum of Law, at pp. 10-11, the Supreme Court's citation, in Teamsters, 431 U.S. at 335 n. 15, 97 S.Ct. 1843, to its previous opinion in Village of Arlington Heights v. Metropolitan Housing Development Corp., 429 U.S. 252, 97 S.Ct. 555, 50 L.Ed.2d 450 (1977), is most instructive with respect to proof of discriminatory motive. In Arlington Heights, the Court noted that, since legislators and administrators are concerned with balancing numerous competing considerations in making decisions, as must the administration of Swarthmore, "courts refrain from reviewing the merits of their decisions, absent a showing of arbitrariness or irrationality." 429 U.S. at 265-266, 97 S.Ct. at 563. This judicial deference is no longer justified "[w]hen there is a proof that a discriminatory purpose has been a motivating factor in the decision. . . ." Id. The Court in Arlington Heights then went on to explain as to how one might establish an invidious discriminatory purpose:

Determining whether invidious discriminatory purpose was a motivating factor demands a sensitive inquiry into such circumstantial and direct evidence of intent as may be available. The impact of the official action — whether it "bears more heavily on one race than another," Washington v. Davis, 426 U.S. [229], at 242 [ 96 S.Ct. 2040, at 2049, 48 L.Ed.2d 597] — may provide an important starting point. Sometimes a clear pattern, unexplainable on grounds other than race, emerges from the effect of the state action even when the governing legislation appears neutral on its face. . . . The evidentiary inquiry is then relatively easy. But such cases are rare. Absent a pattern as stark as that in Gomillion or Yick Wo, impact alone is not determinative, and the Court must look to other evidence.¹⁵ (Citations omitted.)

Accordingly, it is the plaintiff's burden in a Title VII disparate treatment case to establish not only the existence of disparate treatment but also that such treatment was caused by purposeful or intentional discrimination.

Since plaintiffs in effect have brought a pattern and practice case, Teamsters explicitly defined what the plaintiffs must do in order to establish a prima facie case of class discrimination. They must show that alleged differences in treatment between men and women are sexually premised. In addition, since plaintiffs alleged a system-wide pattern or practice of resistance to the full enjoyment of Title VII rights, they must show by a preponderance of the evidence that sex discrimination against female faculty members is the standard operating procedure of Swarthmore. As the Teamsters Court noted:

the Government ultimately had to prove more than the mere occurrence of isolated or "accidental" or sporadic discriminatory acts. It had to establish by a preponderance of the evidence that racial discrimination was the company's standard operating procedure — the regular rather than the usual practice. 431 U.S. at 336, 97 S.Ct. at 1855 (footnote omitted).

One of the most widely used and effective means of establishing a pattern or practice of discrimination is by the use of statistics. As the Court noted both in Teamsters and in Hazelwood School District v. United States, 433 U.S. 299, at 306-309, 97 S.Ct. 2736, at 2741-2742, 53 L.Ed.2d 768 (1977), statistics can be an important source of proof in employment discrimination cases. This is due to the fact that:

absent explanation, it is ordinarily to be expected that nondiscriminatory hiring practices will in time result in a work force more or less representative of the racial and ethnic composition of the population in the community from which employees are hired. Evidence of longlasting and gross disparity between the composition of a work force and that of the general population thus may be significant even though § 703(j) makes clear that Title VII imposes no requirement that a work force mirror the general population. Teamsters, supra, 431 U.S. at 339 n. 20, 97 S.Ct. at 1856-57.

Furthermore, "[w]here gross statistical disparities can be shown, they alone may in a proper case constitute prima facie proof of a pattern or practice of discrimination. Teamsters, supra, 431 U.S. at 339-340 [ 97 S.Ct., at 1856]." Hazelwood, supra, 433 U.S at 307, 97 S.Ct. at 2741. However, the Court in Teamsters went on to state that:

[w]e caution only that statistics are not irrefutable; they come in infinite variety and, like any other kind of evidence, they may be rebutted. In short, their usefulness depends on all of the surrounding facts and circumstances. Teamsters, supra, 431 U.S. at 340, 97 S.Ct. at 1856-57.

The Court also noted that the value of statistics may be lessened where there are "considerations such as small sample size" or there is "evidence showing that the figures for the general population might not accurately reflect the pool of qualified job applicants. . . ." Teamsters, supra, at 340 n. 20, 97 S.Ct. at 1857.

Defendants contend, at pp. 14-16 of their Supplemental Memorandum of Law, that cases such as Teamsters and Hazelwood which make statistical comparisons of a defendant's work force with the available labor pool or relevant labor market are not relevant in evaluating the statistical evidence presented by plaintiffs in this case through Dr. deCani. This is due to the fact that, for the most part, Dr. deCani's statistical studies on rank, hiring, promotion and salary were internal comparisons concerned with analyzing Swarthmore's faculty given the members actually appointed, and determining the probabilities of the distribution of the faculty by rank, length of contract and salary if that distribution were determined by chance alone. The fact that the particular statistical methods utilized in Teamsters and Hazelwood shed little light on plaintiffs' statistical methodology is irrelevant, since Teamsters and Hazelwood explicitly recognized that statistics come in an infinite variety and that their usefulness depends on all of the surrounding facts and circumstances. Therefore, although an infinite variety of statistics may be used in an attempt to establish a prima facie case of a pattern or practice of sex discrimination, the Court must carefully analyze the particular statistical methods used and then, assuming their propriety, must determine whether the statistics, either alone or in conjunction with other evidence, establish that Swarthmore has regularly and routinely discriminated against women faculty members.

In addition to their statistical evidence, plaintiffs chose to bolster their statistical presentation by offering the testimony of individual class members. The defendants' position is that, although at the liability stage the plaintiffs are not required to offer evidence that each person for whom it will ultimately seek relief was a victim of the employer's discriminatory policies, see Teamsters, supra at 324, 97 S.Ct. 1843, if individual cases are used as evidence to create, in conjunction with statistics, an inference of a pattern or practice of discrimination, then the individual cases must meet the formula for establishing a prima facie case of discrimination outlined by the Supreme Court in McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973). On the specific facts there involved, the Court held that this burden may be met by showing that a qualified applicant, who was a member of a racial minority group, had unsuccessfully sought a job for which there was a vacancy and for which the employer continued thereafter to seek applicants with similar qualifications. Upon such a showing, the burden then shifted to the employer to articulate some legitimate, nondiscriminatory reason for the employee's rejection. Id. at 802, 93 S.Ct. 1817. On the other hand, the plaintiffs' position is that whether or not the specific instances meet the McDonnell Douglas test, "they certainly rise to the level of bolstering the statistical evidence such that the evidence as a whole creates an inference of sex discrimination." [N.T. 27-18.]

First of all, if a class member testifies in the liability stage of the class case, his or her testimony need not exactly fit the McDonnell Douglas formula. As the Court pointed out in Teamsters, supra at 324, 97 S.Ct. 1843, their decision in McDonnell Douglas "did not purport to create an inflexible formulation. . . . The importance of McDonnell Douglas lies not in its specification of the discrete elements of proof there required, but in its recognition of the general principle that any Title VII plaintiff must carry the initial burden of offering evidence adequate to create an inference that an employment decision was based on a discriminatory criterion illegal under the Act." Teamsters, supra at 358, 97 S.Ct. at 1866 (footnote omitted).

Second, what an individual class member's burden of proof is depends on whether she testifies and, if she does testify, what her testimony consists of. For example, if plaintiffs are successful in establishing a broad-based policy of employment discrimination through the use of statistics alone, then a court may infer that individual hiring decisions were made in pursuit of the discriminatory policy. Teamsters, supra at 324, 97 S.Ct. 1843. The plaintiffs must then only show that an alleged individual discriminatee unsuccessfully applied for a job, was denied promotion or was not granted tenure, though eligible, and was therefore, a victim of proven discrimination. Id. at 324, 97 S.Ct. 1843. However, if plaintiffs present the testimony of class members to bolster statistical evidence in an attempt to make out a prima facie case, then, whether or not that class member must carry the initial burden of offering evidence adequate to create an inference of unlawful discrimination, depends on the content of the testimony. Thus, if a female faculty member were to testify that she overheard a conversation between the President and the Provost wherein the President of the College told the Provost to make sure that no women are promoted above the rank of Assistant Professor, then such testimony obviously would not have to meet a McDonnell Douglas-type formula. However, if a female faculty member were to testify, as many did in this case, that, by reason of their own personal experience at Swarthmore, they were victims of alleged discrimination, then they would have to show more than the mere fact that they were not hired, promoted, renewed, tenured, etc. They must also demonstrate that the alleged discrimination did not result from the most common legitimate non-sexist reason on which Swarthmore might deny a female faculty member a promotion, renewal and tenure — an absolute or relative lack of qualification — and in the case of hiring, an absence of a vacancy. See Teamsters, supra at 338 n. 19, 358 n. 44, 97 S.Ct. 1843.

The final legal issue we must address is the effect of the Supreme Court's recent decisions in United Air Lines, Inc. v. Evans, 431 U.S. 553, 97 S.Ct. 1885, 52 L.Ed.2d 571 (1977), and in Hazelwood, supra, as to the relevance of pre-Act discrimination which, in this case, are alleged discriminatory acts which took place prior to March 24, 1972 — the date educational institutions became subject to Title VII's coverage.

First, we may quickly dispose of any argument that Evans precludes this Court from considering any act of discrimination which was not itself the subject of a timely charge. [ See N.T. 27-26 to 27-27.] This is due to the fact that we have already decided that Barbara Presseisen, who herself filed timely charges, may properly bring a class action on behalf of those who had not filed charges with the EEOC, 71 F.R.D. 34, 47-48 (E.D.Pa. 1976); see United Airlines, Inc. v. McDonald, 432 U.S. 385, at 394, n. 6, 97 S.Ct. 2464, 53 L.Ed.2d 423 (1977), as long as the class members left the employ of Swarthmore on or subsequent to March 24, 1972, or applied for faculty positions and were rejected on or subsequent to March 24, 1972. The crucial issue then is how do we treat pre-Act discrimination.

As the Court noted in Hazelwood, racial discrimination by public employers, like sex discrimination by educational institutions, was not made illegal until March 24, 1972. Accordingly, a public employer or, as in this case a college, "who from that date forward made all its employment decisions in a wholly nondiscriminatory way would not violate Title VII even if it had formerly maintained an all-white work force by purposefully excluding Negroes." 433 U.S. at 309-310, 97 S.Ct. at 2742 (footnote omitted). In Evans, the Court also noted that:

[a] discriminatory act which is not made the basis for a timely charge is the legal equivalent of a discriminatory act which occurred before the statute was passed. It may constitute relevant background evidence in a proceeding in which the status of a current practice is at issue, but separately considered, it is merely an unfortunate event in history which has no present legal consequences. 431 U.S. at 558, 97 S.Ct. at 1889.

Of course, even though pre-Act discrimination has no present legal consequences, under narrowly defined circumstances, it may have probative force. For example, as pointed out in 433 U.S. at 309 n. 15, 97 S.Ct. at 2742, in Hazelwood:

[p]roof that an employer engaged in racial discrimination prior to the effective date of Title VII might in some circumstances support the inference that such discrimination continued, particularly where relevant aspects of the decision-making process had undergone little change. Cf. Fed. Rule Evid. 406; Village of Arlington Heights v. Metropolitan Development Corp., supra; 2 J. Wigmore, Evidence §§ 92, 302-305, 371, 375 (3d ed. 1940).

Since, for the reasons stated in our findings infra, we do not find that Swarthmore's employment practices before 1972 were so clearly discriminatory, pre-1972 acts of discrimination will be accorded little, if any, evidentiary weight. In addition, we do not believe that any pre-Act individual instance of discrimination, such as the failure of a female faculty member to be promoted from Assistant to Associate Professor in 1968, for example, can be deemed a continuing violation such that relief may be awarded for such an alleged act of discrimination. The act of discrimination took place in 1968 and is in the category of being "merely an unfortunate event in history which has no present legal consequences." But see EEOC; Interpretative Memorandum, released 7/12/77, 46 U.S.L.W. 2028.

Finally, plaintiffs argued for the first time at final arguments [N.T. 27-21 to 27-22] that pre-Act individual discrimination incidents can still serve as a basis for liability, if the discriminatory incident is locked-in by a post-Act sexually neutral employment practice. As mentioned above, these are claims that stress "disparate impact" and involve employment practices that are facially neutral in their treatment of different groups but that in fact fall more harshly on one group than another and cannot be justified by business necessity. Examples of such neutral practices are the use of qualifying tests, see Griggs, supra, or the height and weight requirement for prison guards which disparately exclude women. See Dothard, supra. The only facially neutral lock-in practice identified by the plaintiffs is contained in Proposed Finding of Fact V.B.2, wherein they allege that "annual raises are determined across the board by rank on a percentage basis such as for example, 5% for Professors, 7 1/2% for Associate Professors and 8% for Assistant Professors." Plaintiffs argue, for example, that if a woman were hired initially as an Instructor but should have been initially hired as an Assistant Professor but for sex discrimination, then Swarthmore's annual salary percentage increase has a continuing impact on her pay, i. e., she will never make up the money lost by being hired initially at a lower rank.

The Court has two responses to plaintiffs' contention. The first is that plaintiffs have always, from the beginning of this case until the middle of the 27th day of trial, treated this case as one of disparate treatment and their late-inning attempt to establish a disparate impact theory is somewhat disturbing. Second, Evans held that a sexually neutral seniority system which gives present effect to a past act of discrimination cannot be the basis of any liability if the complainant failed to file a timely charge of discrimination. Since Evans held that a discriminatory act which is not made the basis for a timely charge is the legal equivalent of a discriminatory act which occurred before the statute was passed, it could be held that pre-Act discriminatory incidents, even if locked-in by a post-Act neutral practice, cannot serve as a basis for liability.

In their final argument, plaintiffs sought to distinguish Evans on the ground that it was a seniority case, whereas obviously this case is "a percentage salary increase case." While it is true that Evans does discuss § 703(h) of Title VII, 42 U.S.C. § 2000e-2(h), which provides that it shall not be an unlawful practice to apply different terms of employment pursuant to a bona fide seniority system if any disparity is not the result of intentional discrimination, it can be persuasively argued, as Judge Becker recently did in Freude v. Bell Telephone Co., 438 F. Supp. 1059, (E.D.Pa., 1977), that "it is plain that the basic holding of Evans is that a current nondiscriminatory policy will not revive a time-barred [or, as in this case, a pre-Act] act of discrimination even though such policy has a continuing impact and gives present effect to a past act of discrimination. That holding is not based on the protection afforded a bona fide seniority system by § 703(h) and it applies directly to the instant [pension] case." 438 F. Supp. at 1061. Accord, Dickerson v. United States Steel Corp., 439 F. Supp. 55, 69-71 (E.D.Pa., filed July 25, 1977) (Newcomer, J.).

Were it not for Griggs v. Duke Power Co., 401 U.S. 424, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971), this Court would be quick to hold, on the basis of Evans, that all discriminatorily neutral policies cannot be used to revive past acts of discrimination which are, for whatever reason (pre-Act or time-barred), unfortunate events in history which have no present legal consequences. In Griggs, the Court held that non-job-related requirements as a condition of employment or transfer were, even though facially neutral, prohibited since they operated to exclude Negroes. In other words, Title VII "proscribes not only overt discrimination but also practices that are fair in form, but discriminatory in operation." Griggs, 401 U.S. at 431, 91 S.Ct. at 853. As stated by the Court in Teamsters, "one kind of practice `fair in form, but discriminatory in operation' is that which perpetuates the effects of prior discrimination. As the Court held in Griggs, supra: `Under the Act, practices, procedures, or tests neutral on their face, and even neutral in terms of intent, cannot be maintained if they operate to "freeze" the status quo of prior discriminatory employment practices.' 401 U.S., at 430, 91 S.Ct. [849] at 853." 431 U.S. at 349, 97 S.Ct. at 1861. Most importantly, the prior discriminatory employment practices in Griggs occurred prior to the original enactment of Title VII. See also Teamsters, at 349 n. 32, 97 S.Ct. 1843. Notwithstanding the fact that Teamsters was decided the same day as Evans, the Court went on to point out that, were it not for § 703(h), the Teamsters' seniority system, which unlawfully perpetuated the effects of pre-Act discrimination, "would seem to fall under the Griggs rationale." At 349, 97 S.Ct. at 1862. Finally, it is interesting to note what the Court in Teamsters actually thought was the holding of the Court in Evans. At 348 n. 30, 97 S.Ct. at 1861, of Teamsters, the Court stated:

Evans holds that the operation of a seniority system is not unlawful under Title VII even though it perpetuates post-Act discrimination that has not been the subject of a timely charge by the discriminatee. (Emphasis added.)

We need not finally decide this issue, however, since, as we stated at the outset, plaintiffs have maintained from the beginning that this is a disparate treatment case. We do not believe that one proposed finding of fact out of one hundred and fifty-five pages of proposed findings of fact comes close to establishing a neutral practice which perpetuates alleged past discrimination as to all class members. The Court will leave the above-mentioned Griggs-Teamsters-Evans pre-Act perpetuation problem for another day and another case.

FINDINGS OF FACT — Continued

IV. ORGANIZATION OF THE COLLEGE

The ultimate authority of Swarthmore resides in the Board of Managers, which operates through certain standing committees. The President is the principal administrative officer, responsible to the Board. [N.T. 17-21 to 17-24.] The Vice-President for administration is the Controller responsible directly for financial records and for financial matters. The Provost is the principal academic officer with primary responsibilities in matters relating to faculty, curriculum and library. [N.T. 17-11 to 17-12.] The definition of functions of major officers of the institution and a description of many faculty procedures are set forth in the College faculty handbook. [Ex. D-2.] Department chairmen are appointed by the President of the College. Since 1968, such appointments have been set at five-year terms. [Deposition of Charles Gilbert, at 132-133.] Department chairmen have responsibility for the administration of departmental matters, including such personnel matters as recruitment, hiring, contract renewal, compilation of dossiers for promotion and tenure, recommendations for promotion and tenure, and salaries. [Deposition of Charles Gilbert at 95, 96, 113, 115-119, 136, 138, 142, 143, 153, 154-156; N.T. 18-16, 25, 37, 40.] Department chairmen work with the College Provost in connection with the aforementioned personnel matters. [Ex. P-353 at III-1; N.T. 17-85, 17-89, 18-40; Deposition of Charles Gilbert at 96, 115-118, 142-144, 149-151, 153-157, 169.] Decisions regarding recruitment, hiring, contract renewal and salary are ultimately made by the President and the Provost in consultation with the department chairmen. [Deposition of Charles Gilbert at 95, 96, 113, 115-119, 136, 138, 142, 143, 153-157, 169.] The department chairmen write initial letters of appointment after the appointment and the terms thereof are authorized by the Provost and the President. Said letters are sent to individual faculty members subject to subsequent letters of confirmation of contract terms from the Provost and the President after approval by the College Board of Managers. [N.T. 17-83 to 17-86, 17-89, 17-101; Deposition of Charles Gilbert at 115, 116.] In 1968-69, the College created an advisory Committee on Promotion and Tenure. The committee is generally comprised of four tenured faculty members appointed by the College President. [N.T. 18-28, 18-48.] The President serves as chairman of the committee. [N.T. 18-27.] The Committee on Promotion and Tenure reviews tenure and promotion dossiers that have been compiled by the department chairmen. [N.T. 18-25, 18-26 to 18-29, 18-37, 18-41; Deposition of Charles Gilbert at 170.] Following the deliberations of the Committee on Promotion and Tenure, the President makes a recommendation to the Instructions and Library Committee of the Board of Managers for its action. [N.T. 18-30, 18-31.]

V. INDIVIDUAL NAMED PLAINTIFF — BARBARA PRESSEISEN

Barbara Z. Presseisen was first employed by Swarthmore in May, 1970, as a Lecturer in Education to teach two units in the fall semester beginning September, 1970, and at least one unit in the spring semester beginning January, 1971. Her salary was set at $1,600 per unit plus travel expenses. [Ex. P-1.] At the time of her appointment, she was engaged in writing a dissertation for an Ed.D at Temple University. [N.T. 7-51.] During the 1970-71 academic year, plaintiff became aware of the fact that Swarthmore was searching for a director of the Program in Education. [N.T. 7-98.] Swarthmore was looking for someone who, in addition to being qualified in Education, could function authoritatively as a teacher and scholar in one of the regular departments. [N.T. 24-74 to 24-77.]

From 1951 until May, 1970, when plaintiff was initially hired, Alice Brodhead, who developed the Program in Education, was the only faculty member who devoted herself exclusively to the Program. [N.T. 18-73 to 18-74A.] Mrs. Brodhead was always consulted by Mr. Gilbert before any additional appointments were made in the Program. [N.T. 18-77.]

In June, 1971, plaintiff was appointed as a full-time Assistant Professor of Education for the year 1971-72 at a salary of $10,500. [Ex. P-2] At the time of her appointment, she advised Provost Gilbert that she had almost completed work on her Ed.D degree. [N.T. 7-96.] There is no question that plaintiff was highly qualified for the position of Assistant Professor in the Program. [N.T. 7-73 to 7-76; Ex. P-200A.]

Prior to plaintiff's reappointment in June of 1971, plaintiff met in her office with Provost Gilbert. He informed her that her appointment as an Assistant Professor was for one year only since a successful search for a Director of the Program would preclude her reappointment. [N.T. 23-192 to 23-196; see Ex. P-2; Ex. P-3; Ex. D-66; Ex. D-93.]

Throughout 1970-71 and into the second semester of academic year 1971-72, the relationship between plaintiff and Alice Brodhead was a warm and cordial one. [N.T. 18-77 to 18-79.]

On Saturday, November 13, 1971, Provost Gilbert phoned plaintiff at her home and advised her that she was entitled to a notice of whether she would be reappointed by March 1, 1972, and that, since the search for a Director was continuing, he could not advise her further as to the concrete prospects of reappointment. [N.T. 23-197 to 23-198; Ex. P-4.] This telephone conversation distressed plaintiff, who expressed her concern to Mrs. Brodhead as to the question of reappointment for the 1971-72 academic year. [N.T. 18-78 to 18-79.]

Sometime after the November 13 telephone call, Mrs. Brodhead urged plaintiff to apply for the directorship of the Program and not to decline to do so out of a mistaken sense of loyalty to her (Mrs. Brodhead). However, plaintiff responded that she did not want to apply for the position. [N.T. 18-81 to 18-82.].

On February 28, 1972, plaintiff was advised orally by Provost Gilbert that her one-year contract would not be renewed. [N.T. 7-118, 23-209 to 23-210.] This was confirmed by letter dated February 29, 1972. [Ex. P-5.] Orally, and in the letter, the Provost advised that the nonrenewal was based on logistical considerations; that is, with only two full-time positions which could be allocated to the Program in Education, one position already filled by Mrs. Brodhead, who was reappointed on February 28, 1972, and the other position to be filled by a permanent Director of the Program, there was no position available for Dr. Presseisen. Provost Gilbert also advised Dr. Presseisen that the nonrenewal was not based upon unsatisfactory performance and, in fact, she had performed in a fully satisfactory manner. [N.T. 7-119, 24-80; Ex. P-5.]

At the time of the nonrenewal decision, Swarthmore expected a successful outcome in the search for a Director. [N.T. 23-208; Ex. P-5.] The candidates for the position included two women, Rachel Gelman and Elizabeth Gallistel. [N.T. 23-207 to 23-208.]

The failure to renew Dr. Presseisen was not based on discrimination against her on account of her sex, but was based on the legitimate staffing needs of Swarthmore's Program in Education.

At that time, Dr. Presseisen was not considered for the position of Director not only because she did not apply, but also because she did not meet the requirements for the position. As mentioned above, Swarthmore was looking for someone who, in addition to being qualified in Education, could function authoritatively as a teacher and scholar in one of the regular departments. [N.T. 24-74 to 24-77.]

On March 1, 1972, Dr. Presseisen met for about an hour with President Cross concerning her nonrenewal [Ex. D-75], and by letter dated March 4, 1972, Dr. Presseisen wrote to the American Association of University Professors ("AAUP") contending that the decision not to renew her appointment was based upon improper considerations. A copy of the letter was sent to President Cross. [N.T. 7-130; Ex. P-11.]

In both her meeting with President Cross and in the letter to the AAUP Dr. Presseisen contended that, if Swarthmore could appoint only one person other than the Director, that person should be Dr. Presseisen rather than Mrs. Brodhead, since she was more qualified than Mrs. Brodhead. [N.T. 19-87 to 19-88; Ex. P-11, Ex. D-75.]

Dr. Presseisen revealed to Mrs. Brodhead that she was making such a comparison when Mrs. Brodhead visited her office to correct a misstatement she recalled Dr. Presseisen making, in a previous conversation, with respect to Mrs. Brodhead's qualifications. [N.T. 18-84.] Mrs. Brodhead requested that Dr. Presseisen correct the misstatement of Mrs. Brodhead's qualifications in her letter of March 4, 1972, in which she compared herself favorably with Mrs. Brodhead. [N.T. 18-85 to 18-87; Ex. D-77.] Mrs. Brodhead did not learn of the comparison from anyone other than Dr. Presseisen. [N.T. 18-87, 23-211 to 23-212.]

The above occurrence had a chilling effect on the previously warm and cordial relationship between Mrs. Brodhead and Dr. Presseisen. [N.T. 18-87.] Mrs. Brodhead told Provost Gilbert how upset she was over the situation the same day she had the "comparison" conversation with Dr. Presseisen. [N.T. 18-87 to 18-88.] On the basis of the split between Mrs. Brodhead and Dr. Presseisen, Swarthmore did not consider reappointing Dr. Presseisen when the search for a Director for academic year 1972-73 proved unsuccessful. [N.T. 23-212.] Accordingly, Swarthmore had ample justification for not reappointing Dr. Presseisen when the search proved to be fruitless.

Dr. Presseisen's nonrenewal was reviewed by the faculty members of the Committee on Promotion and Tenure ("Committee"). On March 21, 1972, the Committee found that the nonrenewal was not based upon improper considerations. [Ex. P-14, Ex. P-226.] Pursuant to Dr. Presseisen's request, President Cross, on March 28, 1972, provided her with a written statement of the reasons for her non-reappointment. [Ex. P-15.] Dr. Presseisen did not ask for a further formal review. [N.T. 8-16; Ex. P-227.]

In June of 1972, Swarthmore interviewed Geraldine LaRocque for the position of Assistant Professor of Education. She was interviewed by Mrs. Brodhead, Provost Gilbert and another Swarthmore faculty member, Lee Bramson. She did not receive the appointment, since it was the collective judgment of Provost Gilbert and Mrs. Brodhead that she would not be a suitable appointment. [N.T. 18-89 to 18-91, 23-213 to 23-214.]

In July of 1972, Swarthmore interviewed and hired Robert Pearson for the position previously held by Dr. Presseisen. [N.T. 23-215; Ex. D-83.] Mrs. Brodhead's opinion was that Mr. Pearson was a suitable appointment. [N.T. 18-91.]

Provost Gilbert did not compare the qualifications of Dr. Presseisen with either Geraldine LaRocque or Robert Pearson, since Mr. Gilbert had already decided prior to the interviews of the latter two that Dr. Presseisen was no longer suitable for reappointment. [N.T. 23-215 to 23-216.]

I find, therefore, that the decisions not to reappoint Dr. Presseisen, not to appoint her as the Director, and not to reappoint her once the search for the Director had failed were not made on account of Dr. Presseisen's sex.

I also find that there is no evidence that Dr. Presseisen's salary was discriminatorily set at $10,500, when she was appointed as a full-time Assistant Professor of Education for the year 1971-72.

VI. SCOPE OF THE CLASS

As previously stated, the Court on March 25, 1976, certified the class and defined it as all those women who, on or subsequent to March 24, 1972, have been, are at present or in the future may be: "(1) denied hire, promotion and/or tenure at Swarthmore College or were not recruited by Swarthmore College on account of their sex; and/or (2) denied equal compensation for performance of substantially similar work as performed by male faculty members at Swarthmore College, on account of their sex." As can be seen from our Order, the Court intended that the class be limited to four distinct issues — namely, hiring (which includes recruitment), promotion, tenure (which includes nonrenewal) and salary. Although the Court at trial was very liberal in allowing all types of evidence to be introduced, the Court believes that the class was properly defined with respect to those four issues. Accordingly, notwithstanding the fact that the plaintiffs introduced evidence with respect to a host of other allegedly discriminatory policies on Swarthmore's part, the Court need not address them. For example, at pages 17 through 20 of their Proposed Findings of Fact, plaintiffs discuss part-time people and drill instructors. In addition, at pages 71 through 77 of plaintiffs' Proposed Findings of Fact, plaintiffs also discuss leave time, retirement policies, denial of chairmanships, stipends in aid of publications of dissertations and the denial of a directorship of the art program. The Court holds that such issues are not properly part of the class action, and the Court need not address those issues. The fact that the EEOC was permitted to intervene by this Court on June 4, 1976 (Docket # 85), does not compel us to hold otherwise. An examination of the hearing of the motion of the EEOC to intervene held on June 2, 1976 (Docket # 88, pp. 3-15), and the EEOC's complaint filed on July 14, 1976 (Docket # 93), shows that the EEOC's complaint was and is limited to the issues raised in plaintiffs' original and amended complaints. Since the Court only certified the class as to the above-mentioned categories, the EEOC is bound by those limitations. In conclusion, the Court is not decertifying the class with respect to the above-named categories, such as part-time and drill instructors, but is merely stating for the record that we never certified the class with respect to those issues in the first place. Thus, although the notes of testimony are filled with references to those categories, the Court, for the reasons stated above, will not make findings of fact with respect to those issues.

Upon a review of the evidence, the Court also deems it appropriate to divide the above four issues into sub-classes, since each of those issues represents different groups of women. Accordingly, pursuant to Rule 23(c)(4)(B) of the Federal Rules of Civil Procedure, the Court will divide the class into the following sub-classes: (1) those women who were allegedly discriminated against with respect to tenure, which includes nonrenewals; (2) those women who were allegedly discriminated against with respect to promotion; (3) those women who were allegedly discriminated against with respect to salary; and (4) those women who were allegedly discriminated against with respect to hiring, which includes recruitment.

VII. TENURE

In their attempt to establish a prima facie case that Swarthmore College has a policy of discriminating against women with respect to tenure, the plaintiffs relied on a series of exhibits. Plaintiffs' Exhibit 95 shows that, in 1974-75, 16.7% of the female faculty at Swarthmore was tenured, whereas nationally 34-42% of the female faculty at other colleges was tenured. Plaintiffs' Exhibit 96 shows that, in 1975-76, 12.9% of the female faculty at Swarthmore was tenured, whereas nationally 35-42% of the female faculty at other colleges was tenured. Plaintiffs' Exhibits P-35 and P-35A show that, from 1971 to 1975, the number of tenured females dropped from 6 to 4; the number of tenured males increased from 50 to 68. Plaintiffs' Exhibits P-33 and P-34 show a breakdown by department of the tenured positions in 1971-72 as compared with 1975-76. In 1971, there were 16 departments out of 19 without any tenured women. The only tenured women were in Classics (1), English (1), and Modern Languages (4). In 1975, there were 16 departments without any tenured women. The same three departments had tenured women, but the total number had dropped to four, with Modern Languages losing two tenured women through retirement (Elisa Asensio and Hilda Cohn). Swarthmore's position essentially was that the criteria for appointment to tenure include both the needs of the institution and individual excellence, evidence by scholarship, teaching effectiveness and services to the community as set forth in the faculty handbook. [N.T. 18-22; Ex. D-3, at p. III-6.]

The issue presented then is whether the plaintiffs have established by a preponderance of the evidence that sex discrimination was Swarthmore's standard operating procedure — the regular, rather than the isolated or sporadic incident — with respect to tenure. The Court finds that the plaintiffs have not met that burden. As the plaintiffs themselves admit in their Proposed Findings of Fact IV-B(9), only two women were denied tenure during the period 1966 through 1975: Harriett Shorr and Katherine Morgan. Of those two women, Katherine Morgan was ultimately granted tenure by Swarthmore when she completed the intra-college appeals procedure. We also know that Margaret Anderson was granted tenure in the History department in early 1977. [N.T. 4-139.] Accordingly, even if we were to assume that Harriett Shorr and Katherine Morgan (at least as respects Katherine Morgan in the initial decision not to grant her tenure) were denied tenure because of their sex, the Court still could not find that the Plaintiffs have established that Swarthmore regularly and purposely treated women less favorably than men with respect to tenure.

As was stated by the Supreme Court in International Brotherhood of Teamsters v. United States, supra, 431 U.S. at 360, 97 S.Ct. at 1867, an employer might show that a plaintiff's proof is insignificant because "during the period it is alleged to have pursued a discriminatory policy it made too few employment decisions to justify the inference that it had engaged in a regular practice of discrimination." (Footnote omitted.) See Hazelwood School District v. United States, supra, 433 U.S. at 315, 97 S.Ct. at 2738 n. 1 (Stevens, J., dissenting). Thus, whether or not we conclude that plaintiffs have not made out a prima facie case, because by their own evidence there were only two women denied tenure, or if we conclude that plaintiffs did make out a prima facie case with respect to tenure but that defendants showed that the plaintiffs' proof with respect to tenure was insignificant due to the small number of women denied tenure, the ultimate conclusion is the same. The plaintiffs have not established that Swarthmore regularly and routinely discriminated against women with respect to tenure.

Since the Court has found that plaintiffs have not made out their case with respect to tenure, the Court need not decide the individual cases of Katherine Morgan and Harriett Shorr because neither Ms. Shorr nor Ms. Morgan moved to intervene pursuant to Rule 24 of the Federal Rules of Civil Procedure and are, therefore, not named plaintiffs; and also because, as we have just found, Swarthmore College does not have a policy of discriminating against women on a class-wide basis with respect to tenure. Since plaintiff's complaint with respect to the class only alleges an across-the-board attack with respect to Swarthmore's employment practices and does not affirmatively ask for individual relief on a non-class basis with respect to any other women, the Court is powerless at this point to grant any relief to Ms. Morgan and Ms. Shorr, even if the Court were to conclude that either of these women had been discriminated against on account of sex.

VIII. PROMOTIONS

In their attempt to establish a prima facie case that Swarthmore has a policy of discriminating against women with respect to promotions, the plaintiffs rely upon three types of evidence — statistical evidence, evidence of the promotion procedures and individual examples which seek to bolster the statistical evidence. Before the Court addresses the latter two types of evidence, we believe it is appropriate to discuss initially not only plaintiffs' statistical evidence, but in addition, defendants' statistical evidence and plaintiffs' rebuttal evidence with respect to the statistics.

The ranks of the faculty at Swarthmore from junior to senior are Instructor, Assistant Professor, Associate Professor and Professor. Part-time faculty not appointed to regular rank are designated Lecturer. In their case in chief, the plaintiffs offered the following statistical evidence: (1) The differences between men and women in time from receipt of the highest degree to promotion or appointment to Instructor at Swarthmore, between the 1966-67 and 1975-76 academic years, are as follows: [Ex. P-45A.] The median time for males is 2.5 years; the median time for females is 3 years. The mean time for males is 3.2 years; the mean time for females is 5.7 years. The standard deviation for males is 2.36 years; the standard deviation for females is 5.2 years.

(2) The differences between men and women in time from receipt of the highest degree to promotion or appointment to Assistant Professor at Swarthmore, between 1966-67 and 1975-76, are as follows: [Ex. P-46A.] The median time for males is 3 years; the median time for females is 4 years. The mean time for males is 3.3 years; the mean time for females is 5.8 years. The standard deviation for males is 2.18 years; the standard deviation for females is 4.82 years.

(3) The differences between men and women in time from receipt of highest degree to promotion or appointment to Associate Professor at Swarthmore, between 1966-67 and 1975-76, are as follows: [Ex. P-47A.] The median time for males is 7 years; the median time for females is 9 years. The mean time for males is 7.6 years; the mean time for females is 16.4 years. The standard deviation for males is 3.77 years; the standard deviation for females is 14.03 years.

(4) The differences between men and women in time from receipt of highest degree to promotion or appointment to Professor at Swarthmore, between 1966-67 and 1975-76, are as follows: [Ex. P-48A.] The median time for males is 13 years; the median time for females is 15 years. The mean time for males is 13.9 years; the mean time for females is 14.7 years. The standard deviation for males is 5.54 years; the standard deviation for females is 2.52 years.

Dr. deCani was unable to make any statistical examination of the above data as far as statistical significance is concerned. He did comment, however, that, in his view, the above exhibits demonstrated that, for each rank, the average time to promotion or initial appointment from receipt of highest degree was longer on the average for women than for men on the faculty at Swarthmore, from 1966-67 to 1975-76. [N.T. 12-87, 12-88.]

Before presenting its own statistical evidence with respect to promotions, Swarthmore attempted to show that Dr. deCani's Exhibits P-45A through 48A were unreliable, basically for four different reasons. First, Swarthmore contended that the exclusion from the promotional studies of persons on the faculty prior to 1966, but who were co-faculty members along with persons still on the faculty at the time of, and included in, the promotional studies, has the potential for causing bias in the data concerning relative time to promotion for men and women. These people have repeatedly been referred to as the "inactives." [N.T. 22-17, 22-18.] Second, Swarthmore's expert, Dr. Meier, criticized the promotional studies of Dr. deCani because he believed the variable in question — years since highest degree — "is not a base line characteristic but rather it also has the characteristic of something that may change during the employment and is partly thus a dependent variable rather than an independent variable and analysis which treats it as an independent variable would therefore be not correct." (Emphasis added.) [N.T. 22-24; see also N.T. 22-25.] Third, Dr. deCani's variable (years since highest degree) effectively excludes years of experience prior to receipt of degree. [N.T. 13-101, 13-102, 22-35 to 22-140.] For example, if A, who came to Swarthmore in 1971 with a Masters degree and had 10 years' teaching experience prior to 1971 and subsequently acquired a Ph.D in 1973, was compared with B, who also came to Swarthmore in 1971 with 2 years' prior teaching experience and also acquired a Ph.D in 1973, as A did, Dr. deCani's promotional study would not take into account the fact that A had 8 years' more teaching experience than B did. Finally, Dr. Hollister, who the plaintiffs called in rebuttal, criticized Dr. Iversen's promotional studies which we will discuss infra. However, Dr. Hollister's criticisms of Dr. Iversen's promotional studies apply equally as well to Dr. deCani's promotional studies. For example, Dr. Hollister objected to Dr. Iversen's failure to use control variables in his rank analysis. [N.T. 25-81, 25-82.] Dr. deCani also did not use control factors in his studies of rank. [N.T. 25-114, 26-66.] Dr. Hollister identified the following factors which should be included in a rank analysis: years since degree, division, publications, some assessment of teaching ability, quality of degree, career interruptions, career continuity, quality of publications, administrative responsibility and some measure of committee work. [N.T. 25-81, 25-82, 25-87 to 25-88.]

Swarthmore offered its own statistical studies or promotion through the testimony of Dr. Iversen. His studies show, on their face, that there is no statistically significant or observable difference between men and women in the length of time for promotion from rank to rank and in the length of time within rank at Swarthmore. [N.T. 22-164; Ex. D-59 through D-61] Essentially, D-59, D-60 and D-61 are promotion analyses for Associate Professors, Assistant Professors and Instructors, for those appointed to those ranks in 1966 or later, respectively. Exhibit D-59 shows the mean time from appointment as Associate Professor to promotion to full Professor. For women, it is 5.3 years and for men it is 5.7 years. It also shows the mean time in the rank of Associate Professor for those not yet promoted. For women, it is 3.2 years and for men it is 4.5 years. Exhibit D-60 shows the mean time from appointment as Assistant Professor to promotion to Associate Professor. For women, it is 4.2 years and for men it is 4.5 years. It also shows the length of time as Assistant Professor for those not yet promoted. For women, it is 2.8 years and for men it is 2.9 years. Exhibit D-61 shows the length of time from appointment as Instructor to promotion to Assistant Professor. The mean time for women is 2.2 years and for men it is 2 years. It also shows the mean time as Instructor for those not yet promoted. For women, it is 3 years and for men it is 1.7 years. Accordingly, although plaintiffs' Dr. deCani and defendants' Dr. Iverson both found no statistically significant difference with respect to promotional policies, Dr. deCani found that, on the average, women take longer to get promoted from the date they attain their highest degree; whereas Dr. Iversen concluded that, on the average, women do not take longer to be promoted from rank to rank and, on the average, do not spend a longer period of time within rank.

Plaintiffs then attempted to attack the reliability of Dr. Iversen's rank studies through the cross-examination of Dr. Iversen and through rebuttal testimony of Dr. deCani. In preparing his rank studies, Dr. Iversen believed it was statistically necessary to avoid the inactives problem. Accordingly, he only examined promotion decisions from 1966-67 through 1975-76 and the fall of the 1976-77 academic year. However, as pointed out in his cross-examination at N.T. 23-81 to 23-103, Dr. Iversen left certain people out of his rank studies. First, if a faculty member were promoted to full Professor prior to 1966, he or she was excluded from the rank studies. Second, if a person were last promoted prior to 1966 and still were not promoted as of 1975-76, then they would not be included in the rank studies. Third, if a person were promoted after 1966 and, prior to another promotion, retired prior to 1976-77, then those people would not be included within the rank studies. However, if a person were promoted twice within that 10-year period and retired say in 1975, they would be included within the rank studies. The effect of the method of Dr. Iversen's studies was to exclude 47 members of the faculty who were on the faculty between 1966 and 1976. However, only four of those 47 people were women. [N.T. 26-40.] Dr. deCani then testified in rebuttal with respect to the people that Dr. Iversen excluded in his rank studies. For example, P-636 shows the persons omitted from Dr. Iversen's promotional study. However, P-636 only studies time of promotion and did not study time in rank for those persons still in rank as did the studies of Dr. Iverson. [N.T. 26-36.] In addition, P-636 only deals with full-time people and not regular part-time. Exhibit P-636 shows that the median time for women to be promoted from Assistant Professor to Associate Professor was 6 years; the mean time 7 years. For men the median time was 6 years and the mean time 5.9 years. From Associate Professor to full Professor, the median time for women was 9 years and the median time for men was 6 years. The mean time for women was 10 years and the mean time for men was 7.8 years. Exhibit P-639 shows the persons included in Dr. deCani's promotional studies, and it shows that the median time to promotion from Assistant to Associate Professor for women is 4.5 years and for men it is 5 years; and that the mean time is 4.5 years for women and 4.6 years for men. From Associate to full Professor, the median time for women is 5.5 years and for men 6 years; and the mean time for women is 5.5 years and for men 5.7 years. Exhibit P-641 is a combination of Exhibits P-636, which plaintiffs contend were the people excluded from Dr. Iversen's rank studies, and P-639, which were the people included in his rank studies. However, this again does not include regular part-time people. [ See N.T. 26-37.] Exhibit P-641 shows that, from Assistant to Associate Professor, the median time for women is 5 years and the median time for men is 5 years. The mean time for women is 6.3 years and the mean time for men is 5.4 years. From Associate to full Professor, the median time for women is 9 years and the median time for men is 6 years. The mean time for women is 8.2 years and the mean time for men is 7.2 years. However, Dr. deCani did admit in cross-examination that the differences in median and mean times to promotion between men and women were not statistically significant. In other words, the differences could be based on chance alone. [N.T. 26-55 to 26-56.] Finally, Dr. Hollister testified on behalf of the plaintiffs at N.T. 25-97 to 25-98 that there is really no way to completely eliminate the bias of the inactives problem without introducing other biases.

In addition to their statistical evidence, plaintiffs contend that Swarthmore's criteria for promotion are subjective and vague. There is no question that the criteria for promotion are teaching ability, scholarship appropriate to rank and service to the community. [N.T. 18-41.] Plaintiffs contend that the above standards are vague and not susceptible to reasonable review or evenhanded application. They also contend that the standards for promotion are so vague and the balancing so subjective as to allow for unfettered discretion in any case. Further, in order for an individual to be considered for promotion, such individual must be recommended for promotion by the chairman of the department. [N.T. 18-37.] There is no question that the chairman has wide discretion in determining whether to recommend promotions. Once the chairman recommends an individual for promotion, the Committee on Promotions and Tenure takes up the proposal. The plaintiffs contend that, in making its recommendation to the President to promote or not to promote, the Committee does not record the basis of its recommendation. [N.T. 20-170.] This procedure, the plaintiffs contend, allows for the making of discriminatorily motivated decisions and covers up the actual reasons for the ultimate decision. On the other hand, the defendants contend that the current procedures for promotion are fair and that wide consultation precludes sex biases and influence. [N.T. 18-41 to 18-43, 18-50.] Swarthmore also relies on the fact that since 1968, when the Committee on Promotion and Tenure was created, women have been appointed as members in all but two years — namely, the 1971-72 and 1972-73 academic years. In 1968, two members were women. [N.T. 18-15; Ex. D-15.] Finally, plaintiffs contend at page 60 of their Proposed Findings of Fact that "the statistical disparity in time for promotion for men and women from which discrimination may be inferred is bolstered by the following evidence of individual instances of discrimination in connection with the promotion of female faculty members." (Emphasis added.) The plaintiffs then go on in pages 60 through 70 of their Proposed Findings of Fact to relate six individual instances of discrimination. Essentially, what plaintiffs attempted to do was to compare these six women to what they believed were similarly situated men. From this they argue that these women were discriminated against with respect to promotion. It is important to remind plaintiffs that, in five out of the six comparisons, the non-promotion activity on the part of Swarthmore took place prior to March 24, 1972, the date that Title VII became effective against institutions like Swarthmore. Specifically, plaintiffs complain that Simone Smith was hired as an Instructor and was not appointed to Assistant Professor until 1969-70. They contend that Jean Perkins should have been promoted to full Professor prior to 1971-72. They argue that Hilda Cohn was promoted to Professor in the 1967-68 academic year, despite the fact that she remained an Associate Professor for ten years. They point out that Sue Snyder was hired as an Instructor in 1964 but was not promoted to Assistant Professor until three years later. They tell us that Jean Kopytoff was hired as an Instructor in 1962-63 and it took her three years to become an Assistant Professor. The Court notes that Ms. Kopytoff is not even a member of plaintiffs' class, since she left the employ of Swarthmore well before the enactment of the Equal Employment Opportunity Act of 1972. Finally, they argue that Elisa Asensio, who taught at Swarthmore for 31 years, retired as an Associate Professor and that she should have been promoted to full Professor.

We must keep uppermost in our minds that the issue before the Court is whether the plaintiffs have established a prima facie case that Swarthmore regularly and routinely discriminates against women with respect to promotion. In other words, has Swarthmore had a policy of promoting women much slower than men? The Court does not believe that the plaintiffs have made out a prima facie case. We also find that, even assuming that the plaintiffs had made out a prima facie case with respect to promotion, defendant successfully rebutted that prima facie case.

The Court does not believe that the plaintiffs made out a prima facie case for the following reasons:

First, whether we look at Dr. deCani's original rank studies — namely, P-45A through P-48A — or the rank study he reworked that was offered in rebuttal, Dr. deCani was unable to find any statistically significant disparity in those exhibits. In other words, chance alone could have caused the observed difference in the average times revealed by the exhibits. Thus, the Court believes that on their face plaintiffs' statistical exhibits with respect to promotion are meaningless. They are interesting, but they do not help the Court at all in determining whether Swarthmore has a policy of discriminating against women with respect to promotions. Since we do not believe that the statistics give rise to any inference whatsoever, the subjective procedures and the individual examples really have nothing to bolster. Even if we were to assume that in all six of the individual cases the women were discriminated against with respect to promotion, the Court still does not believe that the plaintiffs have established a prima facie case of discrimination with respect to promotions. As we noted before, five out of the six alleged acts of discrimination took place prior to March 24, 1972. As stated by the Supreme Court in United Air Lines, Inc. v. Evans, supra, 431 U.S. at 558, 97 S.Ct. at 1889, and dealt with more fully in our discussion of the law, supra, discriminatory acts which occurred before the Civil Rights Act was passed "may constitute relevant background evidence in a proceeding in which the status of a current practice is at issue, but separately considered, it is merely an unfortunate event in history which has no present legal consequences." Again, as is dealt with more fully in our discussion of law, supra, the Supreme Court gave us a hint as to what it meant by "relevant background evidence" in Hazelwood School District v. United States, supra, 433 U.S. at 309, 97 S.Ct. at 2742 n. 15. The Court stated:

[t]his is not to say that evidence of pre-Act discrimination can never have any probative force. Proof that an employer engaged in racial discrimination prior to the effective date of Title VII might in some circumstances support the inference that such discrimination continued, particularly the relevant aspects of the decision-making process had undergone little change.

In other words, the Court believes that pre-amendment individual acts of discrimination do not add to plaintiffs' attempt to establish that Swarthmore was discriminating against women with respect to promotions after the amendment. Of course, there could be circumstances where such evidence might be highly relevant. For example, if plaintiffs had shown that, prior to March of 1972, no women had ever been promoted higher than the rank of Assistant Professor, or that prior to 1972 women took so much longer to be promoted from rank to rank or stayed within rank such that a statistician could conclude that the differences have a statistical significance, and then bolstered such statistics with numerous individual acts of discrimination prior to 1972, then the Court might be justified in concluding that such evidence supports the inference that such discrimination continued. However, there was no such evidence presented in this case. Accordingly, we find that the individual cases, even assuming the women were discriminated against, are irrelevant from an evidentiary point of view.

If the plaintiffs had made out a prima facie case with respect to promotions, the Court believes that the defendants adequately rebutted that prima facie case. As we discussed in depth supra, the defendants were able to show that Dr. deCani's promotion studies were unreliable for numerous reasons. That is, the promotion studies do not take into account an individual's prior experience; they do not control for variables, such as division and publications, etc., the fact that the variable in question is not a baseline characteristic, and the inactives problem. The Court finds that those criticisms seriously undercut the reliability of Dr. deCani's promotion studies. In addition, the defendants' own exhibits, prepared by Dr. Iversen, show no observable difference in the length of time for promotion from rank to rank and in the length of time within rank at Swarthmore College between men and women. Granted, there is a problem in Dr. Iversen's exhibits since he, in his attempt to correct for the inactives problem, excluded some people who were on the faculty between 1966-67 and 1975-76. However, the Court does not believe that such omissions undercut the reliability of Dr. Iversen's exhibits since, of the 47 people he excluded from his studies, only 4 were women. Nevertheless, even if we were to find that Dr. Iversen's studies were biased due to the exclusions, then what we are left with is Exhibit P-641, which Dr. deCani prepared for rebuttal. Since he included the people that Dr. Iversen excluded, he put the troublesome inactives problem right back into the formulation. So, what is left is an exhibit that shows a difference between men and women for the time of promotion from Assistant Professor to Associate Professor and for Associate Professor to Professor. But those differences are not great enough to allow it to be said that there is a statistically significant difference and that simply means plaintiffs cannot say that chance alone must be ruled out as the cause of the difference. As we stated with respect to plaintiffs' exhibits P-45A through P-48A, such disparities do not advance the inquiry. The best that can be said about all of the statistical testimony concerning promotions is that, in this case, it is a very difficult area to statistically harness and the various results do not provide a reliable measurement of anything concerning promotions at Swarthmore. Dr. Meier testified, at N.T. 22-60 through 22-62, that the biases may have been eliminated by studying groups of people hired at the same time, possibly in the same division or the same department, and following them through their careers. Although the Court is not qualified to say that such a method may have eliminated all of the biases, it is the Court's view that it may have been a better way to proceed with respect to many of these problems. In conclusion, the Court finds that the plaintiffs have not made out a prima facie case with respect to promotions. That is, the plaintiffs have not shown that Swarthmore regularly and routinely discriminates against women with respect to promotions. Alternatively, even if plaintiffs did make out a prima facie case, the Court finds that the defendants adequately rebutted the prima facie case by showing that plaintiffs' statistical evidence was unreliable and that, for the period 1966 through 1976, there is no statistically significant difference between men and women in the length of time for promotion from rank to rank and in the length of time within rank at Swarthmore.

IX. SALARY

Plaintiffs, in their attempt to establish a prima facie case that Swarthmore has a policy of discriminating against women with respect to salary, relied on statistical testimony and eleven individual cases of alleged discrimination. Defendants attempted to show that plaintiffs' statistics with respect to salary were unreliable and also introduced their own statistics with respect to salary.

Our discussion of the salary issue must, of course, begin with plaintiffs' Exhibit 43A. This exhibit shows that, as Dr. deCani admitted several times in his testimony, once you hold rank constant, there is no significant disparity in the average salaries of men and women. [N.T. 12-73, 12-108 to 12-111, 13-72 to 13-73, 14-48, 14-76 to 14-78.] An examination of P-43A reveals that, except for academic years 1971-72 and 1972-73, at the Assistant Professor level, there is no significant statistical disparity between men and women in the same rank. In other words, any differences between men and women within the same rank could be due to chance alone. Plaintiffs contend, however, and Dr. deCani testified [N.T. 14-76 to 14-78] that, by comparing means and medians, the women for most years made less than men within the same rank. While that is true at the Assistant Professor level, at the Associate Professor level, mean-wise women made more than men four out of five years studied; median-wise, three out of the five years studied; and were equal in one out of the remaining two years. Also, at the Professor level, women had a higher salary median- and mean-wise in one out of the five years studied. [ See also Ex. D-54.] Since plaintiffs were unable to show that Swarthmore had a policy of discriminating against women with respect to salary within rank, plaintiffs attempted to establish a prima facie case of salary discrimination by statistical analysis which excluded rank as a factor. In his multiple-regression analysis (discussed more fully infra), Dr. deCani excluded rank as a variable since, in his view, the inclusion of rank in the salary regression concealed the fact that women at Swarthmore took longer on the average to reach a given rank then did men. This testimony, of course, was based upon Exhibits P-45A through P-48A, which we have thoroughly discussed in the Promotion section. Swarthmore acknowledges that knowing whether women and men were promoted relatively equally with regard to time of promotion is crucial to the salary studies to make certain that there are no differences in the way men and women achieve rank. [N.T. 22-162.] Accordingly, the precise issue before the Court is whether Dr. deCani was properly justified in excluding rank in his multiple-regression analysis. The answer is a simple one, and that is that he was not properly justified in excluding rank. The explanation for our conclusion has already been discussed in detail in the Promotion section. The Court has already found that Swarthmore does not have a policy of regularly and routinely discriminating against women with respect to promotions. Since P-43A shows that women are not discriminated against with respect to salary once you hold rank constant, the plaintiffs have failed to establish a prima facie case that Swarthmore discriminates against women with respect to salary. In addition, even if we were to assume that the eleven individual acts of discrimination (plaintiffs' Proposed Findings of Fact, pp. 42-56) were meritorious, the Court still would not find that plaintiffs have made out a prima facie case because all that you would be left with would be eleven individual examples, since there would be no statistical testimony to buttress.

However, even if Dr. deCani were correct in excluding rank from his multiple-regression analysis, we would still find that plaintiffs have not established by a preponderance of the evidence that Swarthmore discriminated against women with respect to salary.

Dr. deCani testified on the basis of a multiple-regression analysis he performed on the salaries for the full-time faculty at Swarthmore for the years 1971-72 through 1975-76, that sex appeared as a statistically significant factor in each of the five years in the study. [N.T. 13-27 to 13-29; Ex. P-42, revised 3/25/77, hereinafter referred to as P-42.] In the multiple-regression analysis of salaries, Dr. deCani included the variables of sex, age, years since highest degree, years at Swarthmore, degree and division. [N.T. 3-11 to 3-15; Ex. P-42.] Dr. deCani excluded rank as a variable in the regression analysis because, as mentioned above, the inclusion of rank in the salary regression, in his view, concealed the fact that women at Swarthmore College took longer on the average to reach a given rank than did men. [Ex. P-45A through P-48A.] Without going into the intricacies of a statistical definition of exactly what a multiple-regression analysis is, the plaintiffs' multiple-regression analysis of salaries at Swarthmore shows a correlation between the various factors mentioned above, excluding rank, which, when controlled for, shows that sex is a statistically significant factor in the average difference of salaries between men and women. [N.T. 13-48; Ex. P-42.] The chances of a disparity as large, or larger, as that found by Dr. deCani in each of five years studied in P-42, which would be due solely to chance, is .01 for the first three years and .001 for the last two years, or less than one in one-hundred or one in one-thousand, respectively, using a conservative test of significance. From the above, plaintiffs contend that they have shown clearly that sex or a sex-dependent factor was statistically significant in effecting the salaries of men and women on the faculty of Swarthmore from 1971 to 1976. [ See plaintiffs' Proposed Findings of Fact, number 21 at p. 38.] If defendants had not offered any evidence which sought to discredit Dr. deCani's multiple-regression analysis, and if this Court had concluded that Dr. deCani was justified in excluding rank from his analysis, then it is most likely that it would have been appropriate for the Court to find that plaintiffs made out a prima facie case with respect to salary; that is, excluding rank, the evidence would have shown that Swarthmore has regularly and routinely discriminated against women with respect to salary. However, for the reasons stated below, the Court finds that the defendants adequately rebutted such a would-be prima facie case by essentially showing that Dr. deCani's multiple-regression analysis was unreliable. Dr. Meier testified for Swarthmore that the multiple-regression analysis performed by Dr. deCani was unreliable for basically three reasons: First, there is a potential for bias that results from the problem of the inactives which we have already discussed in the Promotion section. Second, there is a potential bias introduced through the aggregation of junior and senior appointments. In other words, Dr. Meier testified that "in an observation study where different strata can be identified, strata that are dealt with differently, it is generally not appropriate to aggregate such strata prior to making an analysis." [N.T. 22-20.] The Court notes that the different strata were identified by Mr. Pagliaro, the present Provost of the College, when he testified (at N.T. 17-91 to 17-93) that, with respect to senior appointments, i. e., Associate Professors and full Professors, the practice of Swarthmore is to generally appoint them for five-year terms; that, with respect to junior appointments, i. e., Instructors and Assistant Professors, the practice has been to appoint them for one-to three-year terms. Dr. Meier went on to explain that the most obvious potential for bias with respect to the aggregation of different strata would arise if the pools from which such candidates are selected are different. He said: "[i]f, for example, it is the case — and I state this hypothetically — that the pool from which senior appointments are drawn is predominantly male and the pool from which junior appointments are drawn is mixed, the result of aggregation will be to mix the effect of seniority with the effect of sex in assessing the average length of initial appointment and make it impossible properly to sort out what policies may be due to seniority and what policies may be due to sex." [N.T. 22-20 to 22-21.] Finally, Dr. Meier testified that, although the regression analysis allows for different intersects, it does not allow for possibly different slopes or different rates of change of salaries in different divisions. [N.T. 22-26 to 22-27.] Dr. Meier went on to explain by illustrative example what he meant by the problem of the non-allowance for possibly different slopes:

Let's suppose that one area, let me choose Physics, is predominantly a male area, that the situation for hiring in Physics and promotion is highly competitive, that is to say that the college competes for a scarce resource and must not only pay high initially but must raise rates rapidly; and contrast that to another area, let us suppose English, where there are a great many candidates, the competition that the college faces for recruitment is not so heavy and they might not raise people as rapidly in that area.

If it were the case that in the departments such as Physics that were male-dominated and in departments such as English that were more mixed, then the differences between policies as regards Physics and English would show up in this analysis as affecting males and females, and that I would describe as a bias in the analysis as given. [N.T. 22-27.]

The final criticism emanates from the testimony of Dr. Hollister, who the plaintiffs called in rebuttal. He testified that Dr. Iversen's regression analysis (D-56) was unreliable since it did not include such factors as scholarship, teaching ability, publications, some assessment of teaching ability, quality of degree, career interruptions, career continuity, quality of publications, administrative responsibility and some measure of committee work. [N.T. 25-87 to 25-88.] As before, it is obvious that the criticisms of Dr. Hollister apply equally as well to the regression analysis of Dr. deCani. [N.T. 13-77 to 13-95; Ex. D-130.] In fact, all parties concede that teaching, publications and research, and service to the community are factors in determining rank and salary at Swarthmore and that there is no way to measure accurately and reliably such data at Swarthmore to include it in a statistical analysis. [N.T. 13-99 to 13-100, 25-91 to 25-92.] The final criticism of Dr. deCani's regression analysis comes not from any testimony offered by any of the parties, but from an observation by the Court. Although there is no question that Dr. deCani is an eminently qualified statistician, we do not believe that his expertise extends into the discipline that governs what goes into the factors to be considered in assessing the amount of an individual's salary at a college. Although Dr. deCani is the chairman of the Statistics Department of the University of Pennsylvania, he was neither called nor qualified as an expert in that area of college administration concerned with the responsibility of the setting of salary. It seems to the Court that it was the plaintiffs' burden to call a witness especially qualified to testify with respect to this area. Thus, plaintiffs could have called an equivalent of the Provost of Swarthmore College from, for example, the University of Pennsylvania or a small liberal arts college like Swarthmore. He or she could have then testified as to what factors typically go into the setting of salaries of a college faculty member. Armed with that type of testimony, Dr. deCani perhaps could have made a more reliable analysis.

The defendants then offered their own evidence through the testimony of Dr. Iversen. He began by stating that the statistical analysis of faculty salaries cannot substitute for a critical analysis of the circumstances of the particular cases, especially at an institution such as Swarthmore where the number of faculty members is so small and the College's need so particularized for special, unique and diverse skills. [N.T. 22-130 to 22-131, 22-134 to 22-135.] With those limitations in mind, Dr. Iversen went ahead and did his own multiple-regression analysis. That analysis is contained in Exhibit D-56, which exhibit is an analysis of faculty salaries of full-time and regular part-time people for the years 1971-72 through the fall of 1976. The variables that Dr. Iversen used in his regression analysis differ somewhat from the variables used in Dr. deCani's. Dr. Iversen included rank, division, sex, years in given rank and degree. Exhibit D-56 shows that the sex variable does not yield a statistically significant disparity. In 1971, there was no measurable difference in the average salaries between men and women. In 1972, women on the average made $340 less than men. In 1973, women made on the average $292 less than men. In 1974, they made on the average $153 less than men. In 1975, women on the average made $303 less than men. In 1976, women on the average made $211 less than men. The P values for those years range from .37 through .66. P values are the means of measuring the probabilities that a given effect differs from zero by chance alone. Since a small value of P, i. e., less than .05 (for example, .04, .03, etc.), indicates an effect is statistically significant; and since all the P values in D-56 are not less than .05, the average differences mentioned above could be attributable to chance alone. In addition, Dr. Iversen conceded that his multiple-regression analysis does not take into account variables such as teaching, scholarship and the rest of the variables that Dr. Hollister testified were important to include in a regression analysis. Also, Dr. Iversen conceded that his regression analysis suffered from the same limitations as expressed by Dr. Meier. [N.T. 22-157.] Finally, Dr. Iversen testified that he made an independent analysis of salaries within divisions, degrees and treating juniors and seniors in different categories, and using as his base point of departure Dr. deCani's regression analysis. He stated:

I spent a considerable amount of time with [Dr. deCani's] equation but realized that it should be modified and I tried following the thought that salaries move differently in the different divisions and for people with different degrees they may not get the same raises and it also struck me that the difference between junior and senior appointments would be important.

So I built these things into the analyses allowing for different slopes and found that the difference between men and women completely disappeared. [N.T. 22-160 to 22-161.]

Plaintiffs did not at any time meet this testimony by Dr. Iversen.

It is true that Dr. deCani attempted to respond to the criticism regarding differences in divisions. Dr. deCani testified with respect to division one as follows:

You must remember the sample signs are very small. The maximum sample signs in division one was eight, which was 1973, '74, and '75. There are eight women. So you are not going to get anything with much significance; but it turns out that in 1973 I got a coefficient for women, which was significant at the five percent level, even with the sample size as small as eight.

There is one other thing. The sign of the coefficient tells you the direction of the disparity. That is the sign negative. That means the expected salary for women is less than the expected salary for a similarly situated male. In all five years the sign of this coefficient was negative.

Now if there were no real differences between men and women you would expect that sign to be positive half the time and negative half the time.

The probability of getting five out of five minuses — if you toss a coin, the probability of getting five out of five heads is one in 32, which is less than .05.

And so on that basis, on the basis of this very simple sign test, there is a significant disparity between the salaries of women and the salaries of similarly situated men in division one. [N.T. 14-82 to 14-85; Ex. P-608 and P-609.]

Dr. deCani did not run a similar test on divisions two and three.

When questioned about the above testimony, Dr. Meier testified that the sign test was an inappropriate calculation. He stated:

[t]he calculation that the probability, if there were no sex bias, of getting the same sign five times in a row assumes, of course, that the observations being referred to are independent. In this case in successive years one is dealing largely with the same set of people.

Thus, if you had a group of individuals and it happened that Group A, the men, were higher paid in 1966 than the women were, it would not be independent judgment that you might expect them also to be higher paid in 1967. These are not independent analyses and there really is no warrant for making the calculation one over two to the fifth power. There are one or two other minor points, but I think that is the major issue, and that calculation is simply inappropriate. [N.T. 22-36 to 22-37.]

Under cross-examination during his rebuttal testimony, Dr. deCani agreed with Dr. Meier's observation:

Let me say the probability was calculated on the assumption that the successive years were independent. Dr. Meier's observation is quite correct. We are dealing with the same group of people in each year and, therefore, the observations in the successive years are not independent. The fact that the sign persisted means that the difference persisted also.

[N.T. 26-28 to 26-69.]

Thus, we can see that Dr. deCani did not adequately meet Dr. Meier's and Dr. Iversen's criticisms with respect to differences within divisions.

Plaintiffs in rebuttal then attempted to discredit Dr. Iversen's regression analysis. First, through the cross-examination of Dr. Iversen, plaintiffs show that Dr. Iversen included in his regression analysis the people that he excluded from his promotion studies. Dr. Iversen admitted that such exclusions and inclusions within the regression analysis had the potential for bias. [N.T. 23-83 to 23-87, 23-106 to 23-110.] Also, Dr. Iversen included regular part-time faculty in his regression analysis. He could not explain, however, the inflation factor that was consistently used to establish each regular part-time faculty member's full-time equivalent salary. [N.T. 23-126 to 23-127.] Plaintiffs contend that, because of the possibility of over-estimation of full-time equivalent salaries for regular part-time faculty, the regression analysis (D-56) is susceptible to considerable bias. In addition to the cross-examination of Dr. Iversen, plaintiffs presented another criticism through the testimony of Dr. Hollister. As mentioned earlier, Dr. Hollister testified that Dr. Iversen should have included more variables than he did in his rank analysis, such as publications and teaching experience. However, as we also stated earlier, those criticisms apply equally to Dr. deCani's regression analysis. Finally, Dr. deCani, while rejecting the potential for bias, during the trial reworked his regression analysis on male and female faculty at Swarthmore for the years 1970 through 1975. [N.T. 26-24 to 26-28.] He attempted to eliminate the inactive bias by excluding from the new regression any people who were on the faculty prior to 1966. He also attempted to eliminate the baseline variable problem by excluding people who had not attained their highest degree by 1971, the first year of the regression study. This new regression study is contained in Exhibit P-642. The results of this reworked regression analysis were very similar to the results of the first regression analysis. [N.T. 26-27.] In each of the five years studied, sex was statistically significant on both P-642 and P-42 and the percentage differences are about the same. [N.T. 26-28.] Thus, Dr. deCani concluded that the inclusion or exclusion of inactives and baseline variables (years since highest degree) produce only a negligible bias. [N.T. 26-28.] Although Dr. deCani attempted to meet Dr. Meier's criticisms with respect to the baseline variables and the inactives problem, Dr. deCani did not meet Dr. Meier's criticisms concerning the differences in division nor did he separate junior from senior appointments, nor did he include the Hollister variables, when he reworked his regression analysis. [N.T. 26-60.]

The Court makes the following observations with respect to the above-described evidence. As we can see from Dr. deCani's regression analysis, to begin with P-42, if you exclude rank, which he did, sex is a statistically significant variable in each of the five years he studied. As mentioned at the outset, if that is where the testimony began and ended, then plaintiffs would have a very strong case. However, as set out above, Dr. deCani's regression analysis is suspect due to the inactives problem, the aggregation of junior and senior appointments, the failure to allow for different slopes within divisions, and the failure to include the Hollister variables. Moving over to Dr. Iversen's regression analysis (D-56), he found that sex was not a statistically significant variable for the years that he studied. Again, as set out above, Dr. Iversen's regression analysis is suspect due to some of the biases brought out by plaintiffs under cross-examination — namely, the inclusion in the regression analysis of persons excluded from his promotional studies, and the lack of explanation with respect to the inflation factor concerning regular part-time faculty members' full-time equivalent salary. Also, the fact that Dr. Iversen admitted that his analysis suffered from the same problems that Dr. Meier pointed out with respect to Dr. deCani's regression analysis. With respect to plaintiffs' Exhibit 642, which is Dr. deCani's reworked regression analysis, there is no question that Dr. deCani attempted to correct for the inactives problem and the baseline variable problem. However, Dr. deCani did not correct for the potential bias introduced by the problem of the aggregation of junior and senior appointments and the different slopes within the division. It seems to the Court that each side has done a superior job in challenging the other's regression analysis, but only a mediocre job in supporting their own. In essence, they have destroyed each other and the Court is, in effect, left with nothing. Again, it is interesting to note that plaintiffs failed to meet Dr. Meier's and Dr. Iversen's criticisms with respect to aggregation of junior and senior appointments and most especially with respect to different slopes within division. As mentioned above, plaintiffs never rebutted Dr. Iversen's testimony at N.T. 22-160 to 22-161 wherein Dr. Iversen stated that he reworked Dr. deCani's original regression allowing for different slopes, and found that the difference between men and women completely disappeared. Perhaps the only conclusion that one can reach, on the basis of all the above-described testimony, is that it is almost impossible, in the fact situation that the Court is presented with, to measure the differences in salaries between men and women by statistical analysis. Finally, if we simply conclude that in any multiple-regression analysis, by definition, there have to be biases, and if, in order to arrive at a result based on the analysis, we simply disregard those biases, what is the Court left with? We are left with Exhibits P-42 and P-642 versus Exhibit D-56. Essentially, the only difference between the two regression analyses is that Dr. deCani excluded rank from his analysis, whereas Dr. Iversen included rank in his analysis. Thus, we have come full circle back to the original problem — whether rank should be included or excluded in the multiple-regression analysis. We have already, in detail, gone through the reasons in the Promotion section why we believe that Swarthmore does not have a policy of discriminating against women with respect to promotion. Since the Court has found that rank, in this particular case, does not conceal more than it reveals about the effect of sex on salary, Dr. deCani improperly excluded rank from his regression analysis. By the same token, Dr. Iversen properly included rank in his regression analysis. Thus, whether the Court finds that Dr. deCani's regression analysis was at the outset improper because it excluded rank; or that the regression analysis is suspect because it does not take into account the criticisms as described in detail above; or whether we find that defendants adequately rebutted plaintiffs' prima facie case with respect to salary; or whether we find that the salary issue is inherently incapable of being analyzed through statistics, the conclusion is the same. Accordingly, the Court finds that Swarthmore does not regularly and routinely discriminate against women with respect to salary. Our conclusion, of course, would not be different even if we were to assume that plaintiffs are correct with respect to the eleven individual acts of discrimination as set forth in their Proposed Findings of Fact at pages 42-56.

Regarding the individual alleged acts of discrimination with respect to salary, some of which suffer from the Evans problem discussed supra, the Court, for the same reasons we expressed in the Tenure section, need not address the merits of those individual claims. Since they have not affirmatively intervened as named party-plaintiffs, and since we have found Swarthmore has no policy of discriminating against women with respect to salary, and since plaintiffs' complaint was limited with respect to that policy question, the Court is again powerless to decide whether Swarthmore discriminated against these women as individuals. Accordingly, we will, because we must, decline to address the merits of those claims.

X. HIRING

A. Overall

In their attempt to establish a prima facie case that Swarthmore discriminates against women with respect to hiring generally, plaintiffs rely on a series of exhibits which they contend show that not enough women have been hired by Swarthmore. First, plaintiffs rely on Exhibits P-49 and P-50. Those exhibits show the number of female and male full-time faculty members between the academic years 1966-47 through 1975-76. We note, however, that P-49 and P-50 include the title Lecturer in the compilations. Since, at trial, it was established that Lecturers are part-time rather than full-time faculty members, the Lecturer category must be excluded and, therefore, the numbers change somewhat. For example, in 1966 the number of full-time female faculty members was 12; in 1971 the number of full-time female faculty members was 18, and in 1975 the number of full-time female faculty members was 22. On the other hand, in 1966 the number of full-time male faculty members was 97; in 1971 the number of full-time male faculty members was 112, and in 1975 the number of full-time male faculty members was 112. In addition to those three years, P-49 shows the following with respect to the number of full-time women on Swarthmore's faculty. In 1966, 12 women; in 1967, 12 women; in 1968, 11 women; in 1969, 12 women; in 1970, 15 women; in 1971, 18 women; in 1972, 18 women; in 1973, 21 women; in 1974, 23 women; and in 1975, 22 women. Thus, plaintiffs' own figures show an increase by almost 50% in the number of full-time female faculty members between the academic years 1966 and 1975, whereas the number of full-time male faculty members remained relatively the same. Specifically, the number of male full-time faculty members remained constant between 1971 and 1975. The Court notes that, even if we include plaintiffs' characterization of Lecturer as full-time, the conclusion is not different. If Lecturer remains in the table, the number of female full-time faculty members in 1966 was 13, and in 1971 it was 21, and in 1975 it was 23; whereas for men it was 101 in 1966, and 113 in 1971, and 112 in 1975.

Another way of looking at the overall statistics is to examine Exhibit P-35 and Exhibits P-49 and P-50 with respect to the percentage of females on the faculty versus the percentage of males on the faculty. The Court has done this by combining the number of full-time faculty members and the number of part-time faculty members. Although there is no class action with respect to part-time faculty members, the Court has chosen to give the following percentages to show that whether or not we examine full-time faculty members only (as discussed above) or whether we examine full-time and part-time faculty together, the conclusion with respect to Swarthmore's overall hiring practices is the same. Thus, in 1966 the percentage of the male faculty was 83.5% and the women faculty was 16.5%; in 1967 for males it was 84.3% and for females it was 15.7%; in 1968 for males it was 81.9% and for females it was 18.1%; in 1969 for males it was 79.7% and for females it was 20.3%; in 1970 for males it was 80.5% and for females it was 19.5%; in 1971 for males it, was 80.8% and for females it was 19.2%; in 1972 for males it was 83.9% and for females it was 16.1%; in 1973 for males it was 81.5% and for females it was 18.5%; in 1974 for males it was 80.2% and for females it was 19.8%; and in 1975 for males it was 80% and for females it was 20%. In addition, defendants' Exhibit D-51 shows the representation of women on the faculty of Swarthmore College; that is, full-time and regular part-time. It shows that, whereas between 1968-69 and 1976-77 when the number of full-time and regular part-time women on the faculty increased from 11 to 27, the number of regular part-time and full-time men only increased from 114 to 117. Accordingly, in 1968-69, the women represented approximately 8.8% of the total number of full-time and regular part-time faculty versus approximately 18.8% in 1976-77.

Accordingly, from an examination of the above exhibits, two things can be seen. First, prior to the enactment of the Equal Employment Opportunity Act of 1972, Swarthmore did not have a history of discriminating against women with respect to hiring. In other words, it is not as if Swarthmore never had a full-time female faculty member prior to 1972. Second, subsequent to 1972, although there was no significant change in the relative number of full-time women on the faculty (that is, from 18 to 22 in the space of four years), neither was there a significant change in the relative number of full-time male faculty during the same time period. In fact, the number of male full-time faculty members remained the same (112). More importantly, any adverse or favorable inference which plaintiffs seek to draw from the above exhibits is limited by the fact that the plaintiffs did not introduce into evidence the relative labor market from which these faculty members were selected. Thus, if plaintiffs had introduced evidence that all other small liberal arts colleges comparable to Swarthmore had a much larger percentage of female faculty members than does Swarthmore, then these numbers might have some probative value. Plaintiffs attempted to do this through Exhibits 95 and 96. However, plaintiffs did not utilize Exhibits 95 and 96 as far as hiring is concerned in their Proposed Findings of Fact. Possibly the reason for this is that, comparing Swarthmore College to private institutional units, other four-year institutional units, private other four-year institutional units, all institutional units and all private institutional units, the percentage of females on Swarthmore's faculty is relatively the same as the percentage of females on the faculties of the above-named institutions. Thus, assuming the validity of Exhibits 95 and 96, and assuming that Swarthmore can appropriately be compared with all of those institutions named above, Exhibit P-95 shows that, for the academic year 1974-75, the percentage of females (full-time and part-time) on Swarthmore's facility was 19.8%, whereas nationwide the percentages ranged from 22% to 24%. Exhibit P-96 shows that, for academic year 1975-76, the percentage of full-time and part-time females on Swarthmore's faculty was 20%, whereas the percentage of females nationwide was 23% to 25%.

Accordingly, whether we look at the raw numbers or the percentage of women on the faculty at Swarthmore, or we compare the percentage of women on the faculty at Swarthmore with other "institutions around the United States," the conclusion is the same. Swarthmore has not historically discriminated against women with respect to hiring, and in fact it has increased by almost 50% the number of full-time women on its faculty between 1966 and 1976. In addition, the number of men has not materially increased and, in fact, between 1971 and 1975, the full-time male faculty members remained the same. In conclusion, the Court simply finds that plaintiffs have not made out a prima facie case that Swarthmore has a policy of discriminating against women with respect to hiring; that is, with respect to hiring the overall faculty.

B. Failure to Hire Women as Associate Professors or Professors

Although we did not find that Swarthmore has a policy of discriminating against women with respect to hiring on an overall basis, the Court is troubled by the fact that no women have ever been hired initially as an Associate Professor or full Professor since 1966. We wish to point out at this time that the plaintiffs are also complaining about the fact that no woman has ever received an initial appointment for longer than three years. However, as we have already mentioned, Mr. Pagliaro testified that Associated Professors or full Professors are typically given an initial appointment for five years, whereas junior appointees such as Assistant Professors and Instructors are typically given one- to three-year appointments. [N.T. 17-91 to 17-93.] Accordingly, it seems to the Court that the two allegations are really the same thing. That is, the reason that no woman was ever appointed for an initial term of longer than three years is due to the fact that no woman was ever initially appointed as an Associate Professor or full Professor. Accordingly, we will discuss the two issues as one under the heading of "Failure to Hire Women as Associate Professors or Professors."

In their attempt to establish a prima facie case with respect to this issue, plaintiffs rely mainly upon Exhibits P-31, P-32, P-32A and P-32B. By way of illustration, P-32A shows rank at first appointment for all persons (full- and part-time) with doctorates between 1966-67 and 1975-76. It shows that between those years, of the 34 women hired by Swarthmore, all of those women were hired at the rank of Assistant Professor or lower. On the other hand, it shows that, out of 190 men hired within that period, 45 were hired initially at the rank of Associate Professor or Professor. Dr. deCani then went on to perform a Wilcoxon rank sum T test on the exhibit. Dr. deCani stated:

Suppose now we draw 34 of these positions at random and assign them to the women. We ask, what is the probability that a random assignment of ranks would produce a distribution as disparate as this in the sense that no women were initially assigned the positions of associate or full professor?

We say, suppose that the assignment of persons to ranks was at random, we drew 34 positions at random and assigned them to the women, what is the probability that the distribution of ranks would be as disparate as this? If the assignment were pure chance, the chances are less than 2 in [100,000]. [N.T. 12-161 to 12-162.]

A similar analysis was applied with respect to Exhibit P-32 which explores the rank at first appointment of full-time faculty members and faculty members with doctorates. Without going through the above explanation again, it shows that, between 1966 and 1975, 12 women were hired and that none of the 12 were hired initially at the rank of Associate Professor or Professor. It also shows that, during the same time period, 135 males were hired and that 32 of those males were initially hired at the rank of Associate Professor or Professor. Since there were only 12 women in the study, the sample size was too small to test for statistical significance. Similarly, Exhibit P-32B shows the rank at first appointment for all persons; that is, full-time and part-time, and persons with or without doctorates. It shows that, during the time period 1966 through 1975, 83 women were hired and that none of those 83 were initially hired at the rank of Associate Professor or Professor. During the same time period, 269 males were hired, and 51 of those 269 were initially appointed at the rank of full Professor or Associate Professor. Dr. deCani testified that, if you drew the 83 positions at random and assigned them to women, the probability that a random assignment of ranks would produce a distribution as disparate as this, in the sense that no woman was initially assigned the position of Associate Professor or full Professor, would be less than 1 in 100,000,000. [N.T. 12-155 to 12-164.] Exhibit P-31 shows the length of initial appointment of full-time people with and without doctorates. It shows that the actual number of females was 36 and that no woman was ever given an initial appointment of longer than three years. On the other hand, out of the 178 men, 32 were given an appointment of longer than three years. Dr. deCani also performed a Wilcoxon rank sum T test on this data and concluded that the probability that no woman would ever receive an initial appointment of longer than three years was less than 2 in 100. [N.T. 13-62 to 13-66.]

Defendants then attacked plaintiffs' Exhibit P-31 through the testimony of Dr. Meier. As mentioned above, he stated that P-31 is insufficient to establish any inference of disparity of treatment between men and women because it inappropriately aggregates junior and senior appointments. [N.T. 22-18 to 22-21.] With respect to Exhibits P-32, P-32A and P-32B, defendants contend they are meaningless because (1) they aggregate inappropriate populations, including those with no title, lecturers and those with regular academic ranks, and (2) they include all those first appointed since 1966 with some, but not all, who were appointed earlier — the so-called inactives problem. [N.T. 22-15 to 22-18.] Finally, it would appear to the Court that Dr. Meier's criticism with respect to the aggregation of junior and senior ranks is equally as applicable to P-32, P-32A and P-32B as it is to P-31. Accordingly, the Court concludes that Exhibits P-31, P-32, P-32A and P-32B are meaningless to the extent that they attempt to utilize the Wilcoxon rank sum T test. However, even assuming that you cannot perform a statistical test on these four exhibits, the exhibits on their face still show the following:

(1) Exhibit P-31 shows that no woman on the faculty between 1966 and 1975 was ever hired for an initial appointment of more than three years at Swarthmore [N.T. 13-63];

(2) No woman on the faculty between 1966 and 1975 was hired for an initial appointment at a rank above Assistant Professor. This was true for part-time as well as full-time positions and whether or not women had Ph.D's [N.T. 12-159, 12-163, 12-164; Ex. P-32, P-32A and P-32B];

(3) From 1966 to 1975, 32 men were hired for initial appointment for longer than three years [Ex. P-31]; and

(4) From 1966 to 1975, 32 men with Ph.D's were appointed full-time initially at ranks of Associate Professor or Professor and 45 men with Ph.D's were appointed full-time and part-time initially at ranks of Associate Professor or Professor. [Ex. P-32, P-32A.] During the same period, 51 men, both with and without doctorates, were appointed initially to the ranks of Associate Professor or full Professor, full- and part-time. [Ex. P-32B.]

As mentioned above, at first the Court sensed discrimination by the fact that no full-time women were ever appointed initially to the rank of Associate Professor or Professor and that no full-time women were ever given an initial appointment of longer than three years. [Of course, we are only concerned with the full-time women and full-time men, since we have already held that there is no class action with respect to part-time persons. Accordingly, we examine only Exhibits P-32 and P-31.] Thus, for example, P-32 shows that between 1966 and 1975, 32 men were originally appointed at the Associate Professor or Professor level, while no women were. On their face, those numbers are somewhat disturbing. However, in order to get a better idea of who those 32 males were and when they were appointed, the Court did an independent analysis of Exhibits P-29 and P-30 with respect to Exhibit P-32. Our analysis revealed the following: First, we found that 35 men were initially appointed at the rank of Associate Professor or Professor, rather than 32 men as stated in Exhibit P-32. Second, we found that 8 of those 35 were initially appointed prior to 1966. The dates range from 1965 all the way back to 1937. In order to be included within that group and in P-29 or P-30, those 8 men, although appointed prior to 1966, had to be on the faculty after 1966. This, of course, is an illustration of the inactives problem. That is, although P-29 or P-30 contain these 8 prior-to-1966 appointees, they do not contain those persons appointed initially to the rank of Associate Professor or full Professor prior to 1966 but who also left the faculty prior to 1966. Third, we found that 18 of those 35 men were initially appointed between the academic years 1966-67 and 1971-72. Two of those 18 were appointed in the 1971-72 academic year. However, since those two persons must have been hired in the year 1971 and could have been hired in the fall of 1970 for the academic year 1971-72, the Court finds that those people must have been hired prior to the date of March 24, 1972. Fourth, that leaves 9 out of the 35 who were hired for the academic years 1972-73 and 1973-74. A breakdown of those 9 shows the following: (1) 6 of those 9 were hired for the 1973-74 academic year in the following departments — 1 in Art, 1 in Engineering, 1 in English, 1 in Modern Languages, 1 in Political Science, and 1 in Religion; (2) the remaining 3 were hired for the 1972-73 academic year in the following departments — 1 in Chemistry, 1 in Math, and 1 in Psychology. Again, since the academic year begins in September, it is possible that those three hired for academic year 1972-73 were hired prior to the enactment of the Equal Employment Opportunity Act on March 24, 1972. Thus, at the outside, plaintiffs have identified only nine men who were hired full-time at the rank of Professor or Associate Professor subsequent to March 24, 1972; and it is possible, for the reasons stated above, that only six were hired subsequent to March 24, 1972. In addition, plaintiffs have not identified any male initially hired at senior ranks after the 1973-74 academic year.

Having broken down Exhibit P-32, the Court finds a number of problems with this presentation by plaintiffs. First and foremost is the fact that, even if P-32 does show a difference, there is no evidence which would show that this difference was caused by sex discrimination. This is due to the fact that plaintiffs did not attempt to show what the relevant labor market for senior appointments was. Thus if plaintiffs had shown that other colleges have a high percentage of women as initial appointees at the senior ranks or that there were a great number of qualified women for those senior positions, then that evidence, in conjunction with the raw numbers on P-32, might be sufficient to raise an inference that the difference was caused by sex discrimination. At that point, it would be the defendants' obligation to come forth and present evidence that, although there were a large number of women qualified for senior appointments, none were available; or that, notwithstanding the fact that there were qualified women, the men were much more qualified in each and every instance. For example, no one could quarrel with the action of a college in an attempt to hire the former Secretary of State as a full Professor, notwithstanding the fact that there are a number of qualified women available to fill the same position. Plaintiffs, of course, rely upon the language in International Brotherhood of Teamsters v. United States, supra. In that case, the Supreme Court implied that proof of gross statistical disparities was, in itself, sufficient to make out a prima facie case of discrimination. See 431 U.S. at 334-343, 97 S.Ct. 1843. Also, in response to defendants' attack on the plaintiffs' statistical showing, the Supreme Court noted that, "[i]n any event, fine tuning of the statistics could not have obscured the glaring absence of minority line drivers. As the Court of Appeals remarked, the company's inability to rebut the inference of discrimination came not from a misuse of statistics but from `the inexorable zero.'" at 342 n. 23, 97 S.Ct. at 1858. The Court does not believe that the above language affords any comfort to the plaintiffs in this case. This is due to the fact, unlike in this case, the Government in International Brotherhood of Teamsters showed that the defendant company hired all white line drivers, notwithstanding the fact that some of the defendant's terminals were in areas of substantial black population. Thus, the Government showed that there was an adequate black labor pool available for the defendant. In this case, the plaintiffs have failed to introduce any evidence of that kind. Without that evidence, neither P-31 nor P-32 gives rise to an inference of discrimination — although those exhibits show a zero, the Court cannot conclude that it is equivalent to the inexorable zero.

As we see it, the problem is that of who has the burden of proof. Plaintiffs' position is that the absence of any initial appointment of women at the rank of Associate Professor or above raises an inference of sex discrimination. The plaintiffs further contend that it is the burden of the College to offer an explanation for the fact that women, especially those with doctorates, were not offered appointments at senior ranks. The Court finds that the College did not have to explain anything, since plaintiffs did not show that there was an available labor pool of women for the senior ranks. Accordingly, we find that the plaintiffs have not made out a prima facie case that Swarthmore has a policy of discriminating against women with respect to senior appointments and length of appointment.

Even assuming that plaintiffs did not have to offer evidence of the relative labor market in order to establish a prima facie case, the Court still would not find that plaintiffs made out a prima facie case since the breakdown of Exhibit P-32 shows that, during the period Swarthmore was alleged to have pursued a discriminatory policy with respect to full-time senior appointments, Swarthmore made too few employment decisions to justify the inference that it engaged in a regular practice of discrimination against women with respect to full-time senior appointments. See International Brotherhood of Teamsters v. United States, supra, at 324, 97 S.Ct. 1843. As mentioned above, Swarthmore initially hired at best nine, and possibly only six, male full-time faculty members at the senior positions subsequent to March 24, 1972. The last of those nine had been hired for September of 1973. This conclusion is not changed by the fact that prior to 1972 Swarthmore initially hired 26 men full-time at senior ranks, whereas they did not initially hire any women, for two reasons: First, 8 of those 26 people were hired prior to 1966, all the way back to 1937. Notwithstanding the inactives problem, the Court finds that such instances are too remote in time to have any probative evidentiary value. Second, with respect to the 18 men who were hired between 1966 and 1971-72, as discussed in Evans such pre-Act alleged discrimination has no present legal consequences. Even assuming that such pre-Act discrimination is a circumstance which might support the inference that such discrimination continued, see Hazelwood School District v. United States, supra, 433 U.S. at 309, 97 S.Ct. at 2742 n. 15, our conclusion would be the same since Swarthmore has made too few employment decisions with respect to senior appointment since the enactment of the Equal Employment Opportunity Act of 1972 for the Court to find that they have a regular practice of discriminating against women with respect to senior appointments.

C. Recruitment

In their attempt to establish a prima facie case with respect to recruitment, plaintiffs rely upon the following evidence. First, Swarthmore had no general policy of advertising vacancies in public journals until the 1974-75 academic year, although there may have been a sporadic use of advertising by individual departments as early as 1965. [N.T. 17-87 to 17-88, 20-126, 21-26 to 21-27.] Second, plaintiffs contend that Swarthmore's procedure in locating applicants to fill vacancies prior to 1975 was to rely upon the "old boy network." This word-of-mouth system, they contend, has a negative effect on the possibility of attracting qualified female applicants. Third, at pages 21 through 25 of plaintiffs' Proposed Findings of Fact, plaintiffs relate individual cases where men were chosen over women to fill particular positions.

Essentially, the defendants' position is that the purpose and effect of the hiring policies and procedures are to meet the need for excellence in their faculty and to obtain a broad spectrum of applicants, thus precluding discrimination on the grounds of sex. Also, Swarthmore contends that the valuation of qualifications and competence of individual candidates is a matter of professional judgment arrived at after a pooling of judgments which should be left to the discretion of Swarthmore, particularly to the members of the department involved and the administration. Finally, Swarthmore contends that the procedures from the time a vacancy is identified and the requirements of the position defined, to the time an appointment is finally made, result in the recruitment of the most qualified candidate, and they are fair to women and men. [See defendants' Proposed Findings of Fact numbers 52, 53, and 55.]

Plaintiffs contend they established a prima facie case of sex discrimination by Swarthmore in their recruitment practices. Our difficulty with this conclusion is twofold. First, the individual cases which plaintiffs identify with respect to recruitment practices are really more properly included in the overall hiring section. That is, these women could not be victims of a sexually discriminatory recruitment policy since, in fact, they were interviewed. Their problem is that they were not hired. Second, the plaintiffs seemingly admit that recruitment is only one of the evidentiary elements which they contend shows that Swarthmore has a pattern or practice of discriminating against women with respect to hiring. For example, in the response of plaintiff and plaintiff-intervenor to defendants' requests for findings of fact and conclusions of law at p. 7, number 65, plaintiffs state that they have shown that men and women have received different treatment, i. e., no women appointed at higher ranks, that the recruitment procedure is subjective and, until 1975, was generally by means of the "old-boy network." This evidence, they contend, is sufficient to give rise to a rebuttable presumption of sex discrimination, and to require defendants to come forth with a non-discriminatory reason for the disparity. Also, in their Proposed Conclusions of Law at pp. 156 and 157, plaintiffs contend that the evidence shows that recruitment was accomplished through word-of-mouth which has resulted in very few women faculty members in many of Swarthmore's departments. The use of such a system, they contend, where it results in exclusion of women, is prohibited by the Act. Thus, we can see that the allegation of allegedly discriminatory recruiting procedures is closely tied to the question of whether Swarthmore has a policy of discriminating against women with respect to hiring. Thus, evidence that the College recruited by the "old-boy network" and that it failed to regularly advertise openings would be highly probative if coupled with evidence that Swarthmore hires very few women to be faculty members. On the other hand, such evidence would be entitled to little weight if, notwithstanding the informal recruiting procedures, the College had a fair representation of women on the faculty. Since we have already concluded, in the previously discussed Hiring Overall section, that Swarthmore does not regularly and routinely discriminate against women with respect to hiring, a fortiori, Swarthmore does not regularly and routinely discriminate against women with respect to recruitment.

XI. REAPPOINTMENTS

In order to establish a prima facie case with respect to non-reappointments, plaintiffs contend that, although the number of women faculty members at Swarthmore is not large, the women who have been on the faculty at Swarthmore are heavily concentrated in certain departments where they experience unusual procedural and substantive obstacles to promotion, tenure, salary increases and reappointment. Further, they contend that, in departments in which there are few or no women, renewal procedures appear to be more regularized; whereas in the departments where women are found in more substantial numbers (Art, Education and Modern Languages), renewal procedures are undertaken on an ad hoe basis. Renewal decisions are based upon evaluation of institutional needs and individual competence, including teaching, scholarship and service to the community. [N.T. 17-104; Ex. D-3, p. III-2A, 3 and 4.] The normal procedures for renewal of appointments include: (1) solicitation of opinions concerning scholarship teaching ability and service to the College from (A) members of the department, (B) colleagues in other departments, (C) students and recent graduates, and (D) scholars in the person's field outside the College; (2) student survey; (3) discussions in the department culminating in the collective judgment of the department; and (4) department recommendation to the President and Provost. [Deposition of Charles Gilbert at pp. 140-142.] Plaintiffs then go on to relate at pp. 26-30 of their Proposed Findings of Fact extremely dissimilar fact situations which they contend establish a pattern and practice of sexually discriminatory reappointment procedures. For example, they point to the fact that Barbara Presseisen was never considered by the College for the position of Director of Education in 1971-72. As we have already found in our findings of fact supra with respect to Barbara Presseisen's individual case, there was no impropriety in Swarthmore's not considering plaintiff for the position of Director. Next, they point to the fact that in March of 1972, while on leave, Harriett Shorr was advised by Provost Gilbert that her appointment would not be renewed. [N.T. 9-126.] However, upon further reflection, Swarthmore ultimately renewed Ms. Shorr. [N.T. 9-126; Ex. P-325A.] Plaintiffs further contend that drill instructors, who are not members of the class, in the modern language departments were hired on a semester basis and generally were not advised whether they would be teaching and how much they would be teaching until the semester was underway and that the drill instructors, most of whom had been women, were frequently paid less salary than what the College had agreed to or raises were not given at all. Finally, the plaintiffs go on to relate a few other incidents which they contend support the notion that Swarthmore has a policy of discriminating against women with respect to reappointment. In their proposed Conclusions of Law, at pp. 159-161, plaintiffs contend that the standard used by the College for renewal are vague, ill-defined and highly subjective, with no written criteria or written reasons preserved to support the decisions in this area; and that, coupled with under-representation of women in the senior ranks, combine to raise the inference of sex discrimination in renewal procedures. In addition, they contend that the department chairmen, who are predominantly male, along with the Provost play a dominant role in evaluating candidates for renewal and tenure and are not given any guidance as to how to apply these highly subjective criteria. As a result, these men are vested with power which results in discriminatory consideration of women for renewal.

The Court does not find that plaintiffs have made out a prima facie case with respect to reappointments for the following reasons:

First, plaintiffs offered no statistics which show that defendants are regularly and routinely discriminatory against women with respect to reappointment. For example, there is no evidence that women on a percentage basis are denied reappointment more frequently than men are, or that Swarthmore has a policy of, for example, when there is a fiscal crunch and faculty members have to go, female faculty members invariably are the first to go. Furthermore, all of the statistical testimony that we set out in detail above belies the fact that Swarthmore has a policy of discriminating against women with respect to tenure, salary, promotion, recruitment, hiring overall or hiring at the senior ranks. In light of those findings, it would be most difficult to find that, notwithstanding that Swarthmore does not discriminate with respect to those categories, they discriminate with respect to reappointment. Finally, a review of the notes of testimony and evidence reveals to the Court that plaintiffs really never attempted to fashion an attack with respect to reappointments as they did with respect to the other categories. This observation is buttressed by the fact that the above examination of plaintiffs' Proposed Findings of Fact shows that their complaints with respect to reappointment are of a scatter-shot variety. Accordingly, the Court finds that plaintiffs have not made out a prima facie case with respect to reappointments.

CONCLUSIONS OF LAW

1. The Court has jurisdiction over the subject matter of this action pursuant to § 706 of the Equal Employment Opportunity Act of 1972, 42 U.S.C. § 2000e et seq. (Supp. V), amending, Civil Rights Act of 1964, 42 U.S.C. § 2000e et seq.

2. The individual plaintiff, Barbara Z. Presseisen, has not established by a preponderance of the evidence that defendants discriminated against her on the basis of sex with respect to any term or condition of employment.

3. The plaintiffs, as class representatives, have not established a prima facie case that defendants regularly and routinely discriminate against female faculty members on account of their sex, with respect to recruitment, hiring and reappointment, on a class-wide basis.

4. The plaintiffs, as class representatives, have not established a prima facie case that defendants regularly and routinely discriminate against female faculty members on account of their sex, with respect to tenure, promotion, and salary, on a class-wide basis. In the alternative, even if plaintiffs, as class representatives, did establish a prima facie case with respect to tenure, promotion and salary, on a class-wide basis, then defendants adequately rebutted plaintiffs' prima facie case.

5. The defendants are entitled to a judgment in their favor and against the individual plaintiff, and against the plaintiff and EEOC as class representatives.

Any factual references under "Discussion" shall be considered as our Findings of Fact in addition to those set forth under "Findings of Fact," supra, and any conclusions of law contained in "Discussion" shall be deemed our Conclusions of Law in addition to those set forth in "Conclusions of Law," supra.

An appropriate Order will be entered.